---------------------------------------------------------------------------------------------------------------------------------------------
      name:  <unnamed>
       log:  C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\Fig3_TabA6_A7.log
  log type:  text
 opened on:   2 Oct 2020, 10:35:17

.   
.   use "E:\ProjData\PlacementCandidacy\utp-DataMainSample.dta", clear

.   
.  // Merge on municipality data 
.    gen kkod = KommunKod

.    
.    capture drop _merge

.    merge m:1 kkod Ar using "E:\ProjData\UtplaceringsData8594\utp-MunData.dta"

    Result                           # of obs.
    -----------------------------------------
    not matched                            37
        from master                         0  (_merge==1)
        from using                         37  (_merge==2)

    matched                           349,383  (_merge==3)
    -----------------------------------------

.    
.   listwise Alder Kvinna Gift_InvAr Barn0_15_InvAr UtbAr_InvAr logpEnklavInvAr Nominerad Vald  InvAlder  InvAr TidI InvAlder logpEnklavVuxen
> InvAr ///
>            mUtb_infodda Herfindahl effectiveparties disproportionality leftshare

.   
.   
.  //################ Main results with municipality controls #################################// 
.   local sample "inrange(InvAr, 1987, 1991) & inrange(InvAlder, 18, 55) & inrange(Ar, 1991, 2014) & FlyktInv == 1 & Rostratt == 1"

.   local controls1 "InvAr IFAUkod Ar FodelseAr"

.   local controls2 "Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr"

.   local controls3 "LoneInk UtbAr Sysselsatt" 

.   local controls4 "RMKvot mUtb_infodda Herfindahl effectiveparties disproportionality leftshare nydrostkval"

.   
.   reghdfe Nominerad logpEnklavVuxen `controls4' if `sample', absorb(KommunValAr `controls1') cluster(KommunValAr) old
(running historical version of reghdfe)
(converged in 9 iterations)

HDFE Linear regression                            Number of obs   =    190,616
Absorbing 5 HDFE groups                           F(   8,    324) =       4.25
Statistics robust to heteroskedasticity           Prob > F        =     0.0001
                                                  R-squared       =     0.0164
                                                  Adj R-squared   =     0.0143
                                                  Within R-sq.    =     0.0005
Number of clusters (KommunValAr) =        325     Root MSE        =     6.4779

                                (Std. Err. adjusted for 325 clusters in KommunValAr)
------------------------------------------------------------------------------------
                   |               Robust
         Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------------+----------------------------------------------------------------
   logpEnklavVuxen |  -.1216476   .0453746    -2.68   0.008    -.2109135   -.0323817
            RMKvot |    1481.06   461.7631     3.21   0.001     572.6279    2389.493
      mUtb_infodda |   -1.37993    .559892    -2.46   0.014    -2.481412   -.2784468
        Herfindahl |  -2.656854   2.350942    -1.13   0.259    -7.281893    1.968184
  effectiveparties |  -.0760899   .1439572    -0.53   0.597    -.3592987     .207119
disproportionality |  -.0516065   .0405178    -1.27   0.204    -.1313177    .0281047
         leftshare |  -.0608749   1.425896    -0.04   0.966    -2.866059    2.744309
       nydrostkval |   .0532591   .0450323     1.18   0.238    -.0353335    .1418517
------------------------------------------------------------------------------------

Absorbed degrees of freedom:
---------------------------------------------------------------------+
       Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
-------------------+-------------------------------------------------|
       KommunValAr |            0             325            325 *   | 
             InvAr |            4               5              1     | 
           IFAUkod |           15              16              1     | 
                Ar |            3               4              1 ?   | 
         FodelseAr |           41              42              1 ?   | 
---------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo c1 

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    190,616    .4275612     6.52484          0        100

.   
.   reghdfe Nominerad `controls4' (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample', absorb(InKommun `controls1') cluster(InKommun) stages
> (first) old 
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    190,616
Absorbing 5 HDFE groups                           F(   8,    285) =     212.99
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.5159
                                                  Adj R-squared   =     0.5150
                                                  Within R-sq.    =     0.2625
Number of clusters (InKommun) =        286        Root MSE        =     0.6110

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .3321209   .0240738    13.80   0.000      .284736    .3795059
              RMKvot |  -56.59817   14.28916    -3.96   0.000    -84.72384    -28.4725
        mUtb_infodda |   .2802314   .0342117     8.19   0.000     .2128917    .3475711
          Herfindahl |  -1.619507   .2458546    -6.59   0.000    -2.103428   -1.135586
    effectiveparties |  -.0865344   .0206186    -4.20   0.000    -.1271185   -.0459504
  disproportionality |   .0969532   .0156838     6.18   0.000     .0660823     .127824
           leftshare |   .4199395   .2167319     1.94   0.054    -.0066589    .8465379
         nydrostkval |   .0262091   .0096114     2.73   0.007     .0072908    .0451275
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            3               4              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   190616
                                                      F(  8,   285) =    16.26
                                                      Prob > F      =   0.0000
Total (centered) SS     =  8043304.263                Centered R2   =   0.0108
Total (uncentered) SS   =  8043304.263                Uncentered R2 =        .
Residual SS             =  8027168.934                Root MSE      =    6.495

------------------------------------------------------------------------------------
                   |               Robust
         Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------------+----------------------------------------------------------------
   logpEnklavVuxen |    -.36166   .1373474    -2.63   0.009     -.632004   -.0913159
            RMKvot |   207.5744   50.47315     4.11   0.000      108.227    306.9219
      mUtb_infodda |   .1030931   .0890367     1.16   0.248    -.0721598     .278346
        Herfindahl |   .6090481   .8720486     0.70   0.485    -1.107425    2.325521
  effectiveparties |   .0373031    .054719     0.68   0.496    -.0704016    .1450077
disproportionality |   .0022785   .0280582     0.08   0.935     -.052949    .0575061
         leftshare |   .9400967   .5411023     1.74   0.083    -.1249672    2.005161
       nydrostkval |   .0003215   .0294016     0.01   0.991    -.0575504    .0581934
------------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             47.251
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              2.4e+04
                         (Kleibergen-Paap rk Wald F statistic):        190.329
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: RMKvot mUtb_infodda Herfindahl effectiveparties
                      disproportionality leftshare nydrostkval
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
---------------------------------------------------------------------+
       Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
-------------------+-------------------------------------------------|
          InKommun |            0             286            286 *   | 
             InvAr |            4               5              1     | 
           IFAUkod |           15              16              1     | 
                Ar |            3               4              1 ?   | 
         FodelseAr |           41              42              1 ?   | 
---------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo c2

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    190,616    .4275612     6.52484          0        100

.    
.   reghdfe Nominerad `controls2' logpEnklavVuxen `controls4' if `sample', absorb(`controls1' KommunValAr) cluster(KommunValAr) old
(running historical version of reghdfe)
(converged in 9 iterations)

HDFE Linear regression                            Number of obs   =    190,616
Absorbing 5 HDFE groups                           F(  12,    324) =       6.32
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.0173
                                                  Adj R-squared   =     0.0153
                                                  Within R-sq.    =     0.0015
Number of clusters (KommunValAr) =        325     Root MSE        =     6.4748

                                (Std. Err. adjusted for 325 clusters in KommunValAr)
------------------------------------------------------------------------------------
                   |               Robust
         Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------------+----------------------------------------------------------------
            Kvinna |  -.1197691   .0354276    -3.38   0.001    -.1894662    -.050072
        Gift_InvAr |   .1119034   .0554748     2.02   0.044     .0027672    .2210395
       UtbAr_InvAr |   .0768131   .0124084     6.19   0.000     .0524019    .1012242
    Barn0_15_InvAr |   .0301393    .018947     1.59   0.113    -.0071354     .067414
   logpEnklavVuxen |  -.1012594   .0443717    -2.28   0.023    -.1885524   -.0139664
            RMKvot |   1483.244   463.9569     3.20   0.002     570.4958    2395.992
      mUtb_infodda |  -1.374436   .5619347    -2.45   0.015    -2.479937   -.2689344
        Herfindahl |  -2.731388    2.35499    -1.16   0.247     -7.36439    1.901615
  effectiveparties |  -.0727943   .1443244    -0.50   0.614    -.3567256     .211137
disproportionality |  -.0502033   .0407594    -1.23   0.219    -.1303897    .0299831
         leftshare |  -.0145375   1.430926    -0.01   0.992    -2.829617    2.800542
       nydrostkval |   .0537467    .045024     1.19   0.233    -.0348297     .142323
------------------------------------------------------------------------------------

Absorbed degrees of freedom:
---------------------------------------------------------------------+
       Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
-------------------+-------------------------------------------------|
             InvAr |            5               5              0     | 
           IFAUkod |           15              16              1     | 
                Ar |            3               4              1 ?   | 
         FodelseAr |           41              42              1 ?   | 
       KommunValAr |            0             325            325 *   | 
---------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo c3

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    190,616    .4275612     6.52484          0        100

.    
.   reghdfe Nominerad `controls2' `controls4' (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample', absorb(InKommun `controls1') cluster(InKo
> mmun) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    190,616
Absorbing 5 HDFE groups                           F(  12,    285) =     171.61
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.5179
                                                  Adj R-squared   =     0.5170
                                                  Within R-sq.    =     0.2656
Number of clusters (InKommun) =        286        Root MSE        =     0.6098

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .3286306   .0239206    13.74   0.000     .2815471    .3757141
              Kvinna |  -.0098817    .005654    -1.75   0.082    -.0210106    .0012472
          Gift_InvAr |   .0189714   .0066482     2.85   0.005     .0058857    .0320571
         UtbAr_InvAr |  -.0125318   .0016751    -7.48   0.000     -.015829   -.0092346
      Barn0_15_InvAr |   .0147224   .0028589     5.15   0.000     .0090952    .0203495
              RMKvot |   -54.8293   14.29122    -3.84   0.000    -82.95903   -26.69956
        mUtb_infodda |   .2895007   .0346574     8.35   0.000     .2212837    .3577177
          Herfindahl |  -1.615691   .2454829    -6.58   0.000    -2.098881   -1.132502
    effectiveparties |  -.0861811     .02081    -4.14   0.000    -.1271418   -.0452204
  disproportionality |   .0970719   .0158506     6.12   0.000     .0658727    .1282711
           leftshare |   .4343306   .2163994     2.01   0.046     .0083867    .8602744
         nydrostkval |   .0263395   .0096049     2.74   0.006     .0074339     .045245
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            3               4              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   190616
                                                      F( 12,   285) =    13.75
                                                      Prob > F      =   0.0000
Total (centered) SS     =  8043304.263                Centered R2   =   0.0119
Total (uncentered) SS   =  8043304.263                Uncentered R2 =        .
Residual SS             =  8018522.838                Root MSE      =    6.492

------------------------------------------------------------------------------------
                   |               Robust
         Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------------+----------------------------------------------------------------
   logpEnklavVuxen |  -.3179111   .1370169    -2.32   0.021    -.5876045   -.0482178
            Kvinna |  -.1073038   .0359864    -2.98   0.003    -.1781366   -.0364709
        Gift_InvAr |   .1031289    .045345     2.27   0.024     .0138753    .1923826
       UtbAr_InvAr |   .0746068   .0102139     7.30   0.000     .0545025    .0947111
    Barn0_15_InvAr |   .0354962   .0179762     1.97   0.049     .0001132    .0708793
            RMKvot |   205.6249   50.55354     4.07   0.000     106.1192    305.1305
      mUtb_infodda |   .0658151   .0902228     0.73   0.466    -.1117725    .2434027
        Herfindahl |   .6888238   .8762982     0.79   0.432    -1.036014    2.413661
  effectiveparties |   .0419521   .0545819     0.77   0.443    -.0654826    .1493869
disproportionality |   .0005267   .0275163     0.02   0.985    -.0536342    .0546876
         leftshare |   .9074718   .5432193     1.67   0.096    -.1617591    1.976703
       nydrostkval |   -.000956   .0294443    -0.03   0.974    -.0589119    .0569998
------------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             47.283
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              2.4e+04
                         (Kleibergen-Paap rk Wald F statistic):        188.743
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr RMKvot
                      mUtb_infodda Herfindahl effectiveparties
                      disproportionality leftshare nydrostkval
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
---------------------------------------------------------------------+
       Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
-------------------+-------------------------------------------------|
          InKommun |            0             286            286 *   | 
             InvAr |            4               5              1     | 
           IFAUkod |           15              16              1     | 
                Ar |            3               4              1 ?   | 
         FodelseAr |           41              42              1 ?   | 
---------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo c4

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    190,616    .4275612     6.52484          0        100

.    
.   esttab c1 c2 c3 c4 using "`output'\Output\MunContTable2.rtf", keep(`controls2' logpEnklavVuxen) ///
>                  replace star(* .1 ** .05 *** .01) b(3) se(3)
(note: file C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\MunContTable2.rtf not found)
(output written to C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\MunContTable2.rtf)

.  
. //######################### Divide the results by left party strength #################################// 
.                                 
. // Regression results Table                             
.   local sample "inrange(InvAr, 1987, 1991) & inrange(InvAlder, 18, 55) & inrange(Ar, 1991, 2014) & FlyktInv == 1 & Rostratt == 1 "

.   local controls1 "InvAr IFAUkod Ar FodelseAr"

.   local controls2 "Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr"

.   local controls3 "LoneInk UtbAr Sysselsatt" 

.   
.   sum leftshare if `sample' , detail

                     (firstnm) leftshare
-------------------------------------------------------------
      Percentiles      Smallest
 1%     .2244482           .116
 5%         .351           .116
10%         .401           .116       Obs             349,383
25%         .438           .116       Sum of Wgt.     349,383

50%         .482                      Mean           .4791737
                        Largest       Std. Dev.      .0784496
75%          .52           .835
90%         .569           .835       Variance       .0061543
95%         .605           .835       Skewness      -.3821891
99%         .673           .835       Kurtosis       5.398655

.   gen StodV = leftshare > r(p50) if leftshare <.

.   
.   reghdfe Nominerad (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample' & StodV ==0, absorb(InKommun `controls1') cluster(InKommun) stages
> (first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    176,209
Absorbing 5 HDFE groups                           F(   1,    284) =     120.80
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4249
                                                  Adj R-squared   =     0.4237
                                                  Within R-sq.    =     0.0549
Number of clusters (InKommun) =        285        Root MSE        =     0.6481

                                     (Std. Err. adjusted for 285 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |    .242921   .0221022    10.99   0.000     .1994162    .2864258
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             285            285 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    285                Number of obs =   176209
                                                      F(  1,   284) =     5.10
                                                      Prob > F      =   0.0247
Total (centered) SS     =  10221423.57                Centered R2   =   0.0098
Total (uncentered) SS   =  10221423.57                Uncentered R2 =        .
Residual SS             =  10207622.56                Root MSE      =    7.619

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.5041348   .2232654    -2.26   0.025    -.9435998   -.0646699
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             46.105
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.0e+04
                         (Kleibergen-Paap rk Wald F statistic):        120.798
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             285            285 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo ca1

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    176,209    .5885057    7.648828          0        100

.   
.   reghdfe Nominerad (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample' & StodV ==1, absorb(InKommun `controls1') cluster(InKommun) stages
> (first) old 
(running historical version of reghdfe)
(converged in 9 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    173,174
Absorbing 5 HDFE groups                           F(   1,    285) =     102.68
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4493
                                                  Adj R-squared   =     0.4482
                                                  Within R-sq.    =     0.0705
Number of clusters (InKommun) =        286        Root MSE        =     0.7006

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .2911573   .0287329    10.13   0.000     .2346018    .3477129
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   173174
                                                      F(  1,   285) =     4.12
                                                      Prob > F      =   0.0432
Total (centered) SS     =  9835982.363                Centered R2   =   0.0119
Total (uncentered) SS   =  9835982.363                Uncentered R2 =        .
Residual SS             =  9824260.864                Root MSE      =     7.54

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.4284184   .2109807    -2.03   0.043    -.8436964   -.0131404
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             28.374
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.3e+04
                         (Kleibergen-Paap rk Wald F statistic):        102.683
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             286            286 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo ca2

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    173,174    .5774539    7.577088          0        100

.   
.   reghdfe Nominerad `controls2' (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample' & StodV ==0, absorb(InKommun `controls1') cluster(InKo
> mmun) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    176,209
Absorbing 5 HDFE groups                           F(   5,    284) =      28.70
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4256
                                                  Adj R-squared   =     0.4244
                                                  Within R-sq.    =     0.0560
Number of clusters (InKommun) =        285        Root MSE        =     0.6477

                                     (Std. Err. adjusted for 285 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .2407763   .0219305    10.98   0.000     .1976093    .2839434
              Kvinna |  -.0148862   .0053719    -2.77   0.006      -.02546   -.0043124
          Gift_InvAr |    .020123   .0086505     2.33   0.021     .0030957    .0371503
         UtbAr_InvAr |   -.007213   .0015451    -4.67   0.000    -.0102542   -.0041717
      Barn0_15_InvAr |   .0049663     .00351     1.41   0.158    -.0019425    .0118751
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             285            285 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    285                Number of obs =   176209
                                                      F(  5,   284) =    12.53
                                                      Prob > F      =   0.0000
Total (centered) SS     =  10221423.57                Centered R2   =   0.0111
Total (uncentered) SS   =  10221423.57                Uncentered R2 =        .
Residual SS             =  10194181.13                Root MSE      =    7.614

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.4264366   .2223803    -1.92   0.056    -.8641593     .011286
         Kvinna |  -.1547598   .0512974    -3.02   0.003    -.2557311   -.0537884
     Gift_InvAr |   .1632013   .0688565     2.37   0.018     .0276673    .2987352
    UtbAr_InvAr |   .1020242   .0140521     7.26   0.000     .0743647    .1296837
 Barn0_15_InvAr |   .0069287   .0248466     0.28   0.781    -.0419782    .0558356
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             46.162
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.0e+04
                         (Kleibergen-Paap rk Wald F statistic):        120.539
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             285            285 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo ca3

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    176,209    .5885057    7.648828          0        100

.   
.   reghdfe Nominerad `controls2' (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample' & StodV ==1, absorb(InKommun `controls1') cluster(InKo
> mmun) stages(first) old
(running historical version of reghdfe)
(converged in 9 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    173,174
Absorbing 5 HDFE groups                           F(   5,    285) =      37.10
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4505
                                                  Adj R-squared   =     0.4494
                                                  Within R-sq.    =     0.0726
Number of clusters (InKommun) =        286        Root MSE        =     0.6999

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .2883597   .0285767    10.09   0.000     .2321115    .3446078
              Kvinna |   -.000597   .0067183    -0.09   0.929    -.0138208    .0126267
          Gift_InvAr |   .0076815   .0095614     0.80   0.422    -.0111384    .0265013
         UtbAr_InvAr |  -.0115046   .0014958    -7.69   0.000    -.0144489   -.0085603
      Barn0_15_InvAr |   .0095205    .003582     2.66   0.008     .0024699    .0165711
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   173174
                                                      F(  5,   285) =    12.29
                                                      Prob > F      =   0.0000
Total (centered) SS     =  9835982.363                Centered R2   =   0.0130
Total (uncentered) SS   =  9835982.363                Uncentered R2 =        .
Residual SS             =  9813014.288                Root MSE      =    7.535

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.3674531   .2123368    -1.73   0.085    -.7854005    .0504943
         Kvinna |  -.1957988   .0561268    -3.49   0.001    -.3062745   -.0853232
     Gift_InvAr |   .1722915   .0799162     2.16   0.032     .0149906    .3295924
    UtbAr_InvAr |   .0882475   .0157994     5.59   0.000     .0571491    .1193458
 Barn0_15_InvAr |  -.0246647   .0277927    -0.89   0.376    -.0793698    .0300403
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             28.359
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.3e+04
                         (Kleibergen-Paap rk Wald F statistic):        101.823
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             286            286 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo ca4

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    173,174    .5774539    7.577088          0        100

.     
.  // Code to significance the difference, use a fully interacted model to replicate the split sample results 
.   gen intT = logpEnklavVuxen*StodV

.   gen intI = logpEnklavVuxenInvAr*StodV

.   
.   reghdfe Nominerad  `controls2' c.StodV#c.Kvinna c.StodV#c.Gift_InvAr c.StodV#c.UtbAr_InvAr c.StodV#c.Barn0_15_InvAr (logpEnklavVuxen intT
>  = logpEnklavVuxenInvAr intI) if `sample', absorb(InKommun##c.StodV InvAr##c.StodV IFAUkod##c.StodV Ar##c.StodV FodelseAr##c.StodV) cluster
> (InKommun) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    349,383
Absorbing 5 HDFE groups                           F(  10,    285) =      29.86
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4396
                                                  Adj R-squared   =     0.4384
                                                  Within R-sq.    =     0.0649
Number of clusters (InKommun) =        286        Root MSE        =     0.6741

                                         (Std. Err. adjusted for 286 clusters in InKommun)
------------------------------------------------------------------------------------------
                         |               Robust
         logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------------------+----------------------------------------------------------------
    logpEnklavVuxenInvAr |   .2407763   .0219396    10.97   0.000     .1975922    .2839605
                    intI |   .0475833   .0314954     1.51   0.132    -.0144099    .1095765
                  Kvinna |  -.0148862   .0053741    -2.77   0.006    -.0254642   -.0043082
              Gift_InvAr |    .020123   .0086541     2.33   0.021      .003089    .0371571
             UtbAr_InvAr |   -.007213   .0015457    -4.67   0.000    -.0102554   -.0041705
          Barn0_15_InvAr |   .0049663   .0035114     1.41   0.158    -.0019453    .0118779
                         |
        c.StodV#c.Kvinna |   .0142892   .0071412     2.00   0.046     .0002329    .0283454
                         |
    c.StodV#c.Gift_InvAr |  -.0124416   .0109857    -1.13   0.258    -.0340649    .0091818
                         |
   c.StodV#c.UtbAr_InvAr |  -.0042917   .0020743    -2.07   0.039    -.0083746   -.0002087
                         |
c.StodV#c.Barn0_15_InvAr |   .0045542   .0044898     1.01   0.311    -.0042832    .0133916
------------------------------------------------------------------------------------------

Absorbed degrees of freedom:
---------------------------------------------------------------------------+
             Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
-------------------------+-------------------------------------------------|
                InKommun |            0             286            286 *   | 
        InKommun#c.StodV |          285             286              1 ?   | 
                   InvAr |            4               5              1     | 
           InvAr#c.StodV |            5               5              0 ?   | 
                 IFAUkod |           15              16              1     | 
         IFAUkod#c.StodV |           16              16              0 ?   | 
                      Ar |            6               7              1 ?   | 
              Ar#c.StodV |            7               7              0 ?   | 
               FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.StodV |           42              42              0 ?   | 
---------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation
WARNING: Missing F statistic (dropped variables due to collinearity or too few clusters).

Stage: first - intT

HDFE Linear regression                            Number of obs   =    349,383
Absorbing 5 HDFE groups                           F(   6,    285) =          .
Statistics robust to heteroskedasticity           Prob > F        =          .
                                                  R-squared       =     0.9586
                                                  Adj R-squared   =     0.9585
                                                  Within R-sq.    =     0.0726
Number of clusters (InKommun) =        286        Root MSE        =     0.4927

                                         (Std. Err. adjusted for 286 clusters in InKommun)
------------------------------------------------------------------------------------------
                         |               Robust
                    intT |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------------------+----------------------------------------------------------------
    logpEnklavVuxenInvAr |  -1.58e-15   1.48e-09    -0.00   1.000    -2.91e-09    2.91e-09
                    intI |   .2883597   .0285885    10.09   0.000     .2320883     .344631
                  Kvinna |   4.49e-17          .        .       .            .           .
              Gift_InvAr |   1.87e-16          .        .       .            .           .
             UtbAr_InvAr |  -1.26e-16   7.06e-11    -0.00   1.000    -1.39e-10    1.39e-10
          Barn0_15_InvAr |  -2.27e-16          .        .       .            .           .
                         |
        c.StodV#c.Kvinna |   -.000597   .0067211    -0.09   0.929    -.0138262    .0126322
                         |
    c.StodV#c.Gift_InvAr |   .0076815   .0095653     0.80   0.423    -.0111462    .0265091
                         |
   c.StodV#c.UtbAr_InvAr |  -.0115046   .0014964    -7.69   0.000    -.0144501   -.0085591
                         |
c.StodV#c.Barn0_15_InvAr |   .0095205   .0035835     2.66   0.008      .002467     .016574
------------------------------------------------------------------------------------------

Absorbed degrees of freedom:
---------------------------------------------------------------------------+
             Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
-------------------------+-------------------------------------------------|
                InKommun |            0             286            286 *   | 
        InKommun#c.StodV |          285             286              1 ?   | 
                   InvAr |            4               5              1     | 
           InvAr#c.StodV |            5               5              0 ?   | 
                 IFAUkod |           15              16              1     | 
         IFAUkod#c.StodV |           16              16              0 ?   | 
                      Ar |            6               7              1 ?   | 
              Ar#c.StodV |            7               7              0 ?   | 
               FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.StodV |           42              42              0 ?   | 
---------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   349383
                                                      F( 10,   285) =     9.73
                                                      Prob > F      =   0.0000
Total (centered) SS     =  20057405.93                Centered R2   =   0.0121
Total (uncentered) SS   =  20057405.93                Uncentered R2 =        .
Residual SS             =  20007195.41                Root MSE      =    7.575

------------------------------------------------------------------------------------------
                         |               Robust
               Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------------------+----------------------------------------------------------------
         logpEnklavVuxen |  -.4264366   .2224717    -1.92   0.056    -.8643326    .0114594
                    intT |   .0589835   .2835956     0.21   0.835    -.4992242    .6171912
                  Kvinna |  -.1547598   .0513185    -3.02   0.003    -.2557711   -.0537484
              Gift_InvAr |   .1632013   .0688848     2.37   0.018     .0276137    .2987889
             UtbAr_InvAr |   .1020242   .0140579     7.26   0.000     .0743538    .1296947
          Barn0_15_InvAr |   .0069287   .0248568     0.28   0.781    -.0419975     .055855
                         |
        c.StodV#c.Kvinna |  -.0410391   .0704114    -0.58   0.560    -.1796313    .0975532
                         |
    c.StodV#c.Gift_InvAr |   .0090902   .0982695     0.09   0.926    -.1843358    .2025163
                         |
   c.StodV#c.UtbAr_InvAr |  -.0137768   .0172716    -0.80   0.426    -.0477728    .0202192
                         |
c.StodV#c.Barn0_15_InvAr |  -.0315934   .0355887    -0.89   0.375    -.1016436    .0384567
------------------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             46.162
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):             9924.930
                         (Kleibergen-Paap rk Wald F statistic):         60.220
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen intT
Included instruments: Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
                      c.StodV#c.Kvinna c.StodV#c.Gift_InvAr
                      c.StodV#c.UtbAr_InvAr c.StodV#c.Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr intI
------------------------------------------------------------------------------

Absorbed degrees of freedom:
---------------------------------------------------------------------------+
             Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
-------------------------+-------------------------------------------------|
                InKommun |            0             286            286 *   | 
        InKommun#c.StodV |          285             286              1 ?   | 
                   InvAr |            4               5              1     | 
           InvAr#c.StodV |            5               5              0 ?   | 
                 IFAUkod |           15              16              1     | 
         IFAUkod#c.StodV |           16              16              0 ?   | 
                      Ar |            6               7              1 ?   | 
              Ar#c.StodV |            7               7              0 ?   | 
               FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.StodV |           42              42              0 ?   | 
---------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo ca1

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    349,383    .5830278    7.613345          0        100

.   
.   lincom _b[intT]

 ( 1)  intT = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |   .0589835   .2835956     0.21   0.835    -.4992242    .6171912
------------------------------------------------------------------------------

.   lincom _b[logpEnklavVuxen]

 ( 1)  logpEnklavVuxen = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |  -.4264366   .2224717    -1.92   0.056    -.8643326    .0114594
------------------------------------------------------------------------------

.   lincom _b[logpEnklavVuxen]+_b[intT]

 ( 1)  logpEnklavVuxen + intT = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |  -.3674531   .2124247    -1.73   0.085    -.7855734    .0506672
------------------------------------------------------------------------------

.   
.   
.   esttab ca1 ca2 ca3 ca4 using "`output'\Output\TableA7_A.rtf", keep(`controls2' logpEnklavVuxen) ///
>                  replace star(* .1 ** .05 *** .01) b(3) se(3)
(note: file C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\TableA7_A.rtf not found)
(output written to C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\TableA7_A.rtf)

.   
.  //######################### Divide the results by the size of NyD in 1991 #################################// 
.                         
. // Regression results Table                             
.   local sample "inrange(InvAr, 1987, 1991) & inrange(InvAlder, 18, 55) & inrange(Ar, 1991, 2014) & FlyktInv == 1 & Rostratt == 1 "

.   local controls1 "InvAr IFAUkod Ar FodelseAr"

.   local controls2 "Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr"

.   local controls3 "LoneInk UtbAr Sysselsatt" 

.   
.   gen aStodNyD = (nydrostkval > 4.4) if nydrostkval <. & Ar == 1991 & `sample'
(330,302 missing values generated)

.   bysort kkod: egen StodNyD = max(aStodNyD)
(20332 missing values generated)

.   
.   reghdfe Nominerad (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample' & StodNyD == 0, absorb(InKommun `controls1') cluster(InKommun) sta
> ges(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    143,782
Absorbing 5 HDFE groups                           F(   1,    285) =     119.18
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4295
                                                  Adj R-squared   =     0.4281
                                                  Within R-sq.    =     0.0747
Number of clusters (InKommun) =        286        Root MSE        =     0.7616

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .3178675   .0291168    10.92   0.000     .2605563    .3751787
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   143782
                                                      F(  1,   285) =     6.31
                                                      Prob > F      =   0.0126
Total (centered) SS     =  10635145.53                Centered R2   =   0.0146
Total (uncentered) SS   =  10635145.53                Uncentered R2 =        .
Residual SS             =  10621056.64                Root MSE      =    8.605

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.5466222   .2176835    -2.51   0.013    -.9750935   -.1181509
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             29.500
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.2e+04
                         (Kleibergen-Paap rk Wald F statistic):        119.181
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             286            286 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo cb1

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    143,782    .7553101    8.658004          0        100

.   
.   reghdfe Nominerad (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample' & StodNyD == 1, absorb(InKommun `controls1') cluster(InKommun) sta
> ges(first) old 
(running historical version of reghdfe)
(dropped 1 singleton observations)
(converged in 9 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    185,268
Absorbing 5 HDFE groups                           F(   1,    282) =      43.48
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4983
                                                  Adj R-squared   =     0.4974
                                                  Within R-sq.    =     0.0591
Number of clusters (InKommun) =        283        Root MSE        =     0.5487

                                     (Std. Err. adjusted for 283 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .2210267   .0335206     6.59   0.000     .1550444    .2870091
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             283            283 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    283                Number of obs =   185268
                                                      F(  1,   282) =     1.37
                                                      Prob > F      =   0.2426
Total (centered) SS     =  7000428.547                Centered R2   =   0.0067
Total (uncentered) SS   =  7000428.547                Uncentered R2 =        .
Residual SS             =  6995955.718                Root MSE      =    6.151

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |   -.249244    .212852    -1.17   0.243    -.6682245    .1697364
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             19.021
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.2e+04
                         (Kleibergen-Paap rk Wald F statistic):         43.478
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             283            283 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo cb2

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    185,268    .3816093    6.165672          0        100

.   
.   reghdfe Nominerad `controls2' (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample' & StodNyD == 0, absorb(InKommun `controls1') cluster(I
> nKommun) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    143,782
Absorbing 5 HDFE groups                           F(   5,    285) =      48.20
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4314
                                                  Adj R-squared   =     0.4300
                                                  Within R-sq.    =     0.0777
Number of clusters (InKommun) =        286        Root MSE        =     0.7604

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .3131995   .0288671    10.85   0.000     .2563798    .3700192
              Kvinna |   -.007298   .0079779    -0.91   0.361    -.0230011     .008405
          Gift_InvAr |   .0307752   .0130107     2.37   0.019      .005166    .0563844
         UtbAr_InvAr |  -.0115225   .0020149    -5.72   0.000    -.0154886   -.0075564
      Barn0_15_InvAr |   .0193287   .0047862     4.04   0.000      .009908    .0287494
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   143782
                                                      F(  5,   285) =    12.20
                                                      Prob > F      =   0.0000
Total (centered) SS     =  10635145.53                Centered R2   =   0.0161
Total (uncentered) SS   =  10635145.53                Uncentered R2 =        .
Residual SS             =  10604792.91                Root MSE      =    8.599

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.4438094   .2179672    -2.04   0.043    -.8728393   -.0147796
         Kvinna |  -.2351214   .0660177    -3.56   0.000    -.3650655   -.1051773
     Gift_InvAr |   .1432037   .1046661     1.37   0.172     -.062813    .3492204
    UtbAr_InvAr |   .1158753   .0201179     5.76   0.000     .0762767    .1554739
 Barn0_15_InvAr |  -.0199948   .0331051    -0.60   0.546    -.0851564    .0451668
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             29.526
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.1e+04
                         (Kleibergen-Paap rk Wald F statistic):        117.716
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             286            286 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo cb3

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    143,782    .7553101    8.658004          0        100

.   
.   reghdfe Nominerad `controls2' (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample' & StodNyD == 1, absorb(InKommun `controls1') cluster(I
> nKommun) stages(first) old
(running historical version of reghdfe)
(dropped 1 singleton observations)
(converged in 9 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    185,268
Absorbing 5 HDFE groups                           F(   5,    282) =      16.96
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4988
                                                  Adj R-squared   =     0.4978
                                                  Within R-sq.    =     0.0599
Number of clusters (InKommun) =        283        Root MSE        =     0.5485

                                     (Std. Err. adjusted for 283 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .2199887   .0334013     6.59   0.000     .1542411    .2857362
              Kvinna |  -.0045238   .0059183    -0.76   0.445    -.0161734    .0071258
          Gift_InvAr |   .0011928   .0074816     0.16   0.873     -.013534    .0159197
         UtbAr_InvAr |   -.006734   .0011301    -5.96   0.000    -.0089586   -.0045095
      Barn0_15_InvAr |  -.0049095    .003114    -1.58   0.116    -.0110391    .0012201
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             283            283 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    283                Number of obs =   185268
                                                      F(  5,   282) =     7.90
                                                      Prob > F      =   0.0000
Total (centered) SS     =  7000428.547                Centered R2   =   0.0075
Total (uncentered) SS   =  7000428.547                Uncentered R2 =        .
Residual SS             =  6990154.274                Root MSE      =    6.148

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |   -.212861   .2157474    -0.99   0.325    -.6375407    .2118188
         Kvinna |  -.1083327   .0497788    -2.18   0.030    -.2063178   -.0103475
     Gift_InvAr |   .1185332   .0550238     2.15   0.032     .0102238    .2268427
    UtbAr_InvAr |    .067518   .0129164     5.23   0.000     .0420932    .0929428
 Barn0_15_InvAr |   .0206814   .0242809     0.85   0.395    -.0271134    .0684762
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             19.031
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.2e+04
                         (Kleibergen-Paap rk Wald F statistic):         43.378
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             283            283 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo cb4

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    185,268    .3816093    6.165672          0        100

.     
.    // Code to significance the difference, use a fully interacted model to replicate the split sample results 
.   capture drop intT intI 

.   gen intT = logpEnklavVuxen*StodNyD
(20,332 missing values generated)

.   gen intI = logpEnklavVuxenInvAr*StodNyD
(20,332 missing values generated)

.   
.   reghdfe Nominerad  `controls2' c.StodNyD#c.Kvinna c.StodNyD#c.Gift_InvAr c.StodNyD#c.UtbAr_InvAr c.StodNyD#c.Barn0_15_InvAr (logpEnklavVu
> xen intT = logpEnklavVuxenInvAr intI) if `sample', absorb(InKommun##c.StodNyD InvAr##c.StodNyD IFAUkod##c.StodNyD Ar##c.StodNyD FodelseAr##
> c.StodNyD) cluster(InKommun) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    329,051
Absorbing 5 HDFE groups                           F(  10,    285) =      38.66
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4645
                                                  Adj R-squared   =     0.4633
                                                  Within R-sq.    =     0.0706
Number of clusters (InKommun) =        286        Root MSE        =     0.6496

                                           (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------------
                           |               Robust
           logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------------+----------------------------------------------------------------
      logpEnklavVuxenInvAr |   .3131995   .0288789    10.85   0.000     .2563566    .3700424
                      intI |  -.0932108   .0463729    -2.01   0.045    -.1844877    -.001934
                    Kvinna |   -.007298   .0079812    -0.91   0.361    -.0230075    .0084115
                Gift_InvAr |   .0307752    .013016     2.36   0.019     .0051555    .0563949
               UtbAr_InvAr |  -.0115225   .0020158    -5.72   0.000    -.0154902   -.0075548
            Barn0_15_InvAr |   .0193287   .0047881     4.04   0.000     .0099041    .0287533
                           |
        c.StodNyD#c.Kvinna |   .0027742   .0092556     0.30   0.765    -.0154438    .0209921
                           |
    c.StodNyD#c.Gift_InvAr |  -.0295824   .0142775    -2.07   0.039    -.0576851   -.0014796
                           |
   c.StodNyD#c.UtbAr_InvAr |   .0047885   .0024303     1.97   0.050     4.91e-06     .009572
                           |
c.StodNyD#c.Barn0_15_InvAr |  -.0242382   .0058023    -4.18   0.000     -.035659   -.0128174
--------------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------------+
               Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------------+-------------------------------------------------|
                  InKommun |            0             286            286 *   | 
        InKommun#c.StodNyD |          284             286              2 ?   | 
                     InvAr |            4               5              1     | 
           InvAr#c.StodNyD |            5               5              0 ?   | 
                   IFAUkod |           15              16              1     | 
         IFAUkod#c.StodNyD |           16              16              0 ?   | 
                        Ar |            6               7              1 ?   | 
              Ar#c.StodNyD |            7               7              0 ?   | 
                 FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.StodNyD |           42              42              0 ?   | 
-----------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation
WARNING: Missing F statistic (dropped variables due to collinearity or too few clusters).

Stage: first - intT

HDFE Linear regression                            Number of obs   =    329,051
Absorbing 5 HDFE groups                           F(   6,    285) =          .
Statistics robust to heteroskedasticity           Prob > F        =          .
                                                  R-squared       =     0.9686
                                                  Adj R-squared   =     0.9685
                                                  Within R-sq.    =     0.0599
Number of clusters (InKommun) =        286        Root MSE        =     0.4116

                                           (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------------
                           |               Robust
                      intT |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------------+----------------------------------------------------------------
      logpEnklavVuxenInvAr |  -1.36e-16   1.14e-09    -0.00   1.000    -2.25e-09    2.25e-09
                      intI |   .2199887   .0334162     6.58   0.000     .1542149    .2857624
                    Kvinna |   9.10e-17          .        .       .            .           .
                Gift_InvAr |   4.90e-17   1.29e-10     0.00   1.000    -2.54e-10    2.54e-10
               UtbAr_InvAr |   5.64e-17   8.37e-11     0.00   1.000    -1.65e-10    1.65e-10
            Barn0_15_InvAr |  -5.51e-17          .        .       .            .           .
                           |
        c.StodNyD#c.Kvinna |  -.0045238   .0059209    -0.76   0.445    -.0161781    .0071304
                           |
    c.StodNyD#c.Gift_InvAr |   .0011928   .0074849     0.16   0.873    -.0135399    .0159256
                           |
   c.StodNyD#c.UtbAr_InvAr |   -.006734   .0011306    -5.96   0.000    -.0089595   -.0045086
                           |
c.StodNyD#c.Barn0_15_InvAr |  -.0049095   .0031154    -1.58   0.116    -.0110415    .0012225
--------------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------------+
               Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------------+-------------------------------------------------|
                  InKommun |            0             286            286 *   | 
        InKommun#c.StodNyD |          284             286              2 ?   | 
                     InvAr |            4               5              1     | 
           InvAr#c.StodNyD |            5               5              0 ?   | 
                   IFAUkod |           15              16              1     | 
         IFAUkod#c.StodNyD |           16              16              0 ?   | 
                        Ar |            6               7              1 ?   | 
              Ar#c.StodNyD |            7               7              0 ?   | 
                 FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.StodNyD |           42              42              0 ?   | 
-----------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   329051
                                                      F( 10,   285) =     9.96
                                                      Prob > F      =   0.0000
Total (centered) SS     =  17635574.07                Centered R2   =   0.0133
Total (uncentered) SS   =  17635574.07                Uncentered R2 =        .
Residual SS             =  17594947.18                Root MSE      =     7.32

--------------------------------------------------------------------------------------------
                           |               Robust
                 Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------------+----------------------------------------------------------------
           logpEnklavVuxen |  -.4438094   .2180564    -2.04   0.043    -.8730147   -.0146041
                      intT |   .2309485   .2870416     0.80   0.422    -.3340421     .795939
                    Kvinna |  -.2351214   .0660447    -3.56   0.000    -.3651186   -.1051241
                Gift_InvAr |   .1432037   .1047089     1.37   0.173    -.0628972    .3493046
               UtbAr_InvAr |   .1158753   .0201262     5.76   0.000     .0762605      .15549
            Barn0_15_InvAr |  -.0199948   .0331187    -0.60   0.547     -.085183    .0451935
                           |
        c.StodNyD#c.Kvinna |   .1267887   .0848542     1.49   0.136    -.0402317    .2938091
                           |
    c.StodNyD#c.Gift_InvAr |  -.0246705   .1208021    -0.20   0.838    -.2624479     .213107
                           |
   c.StodNyD#c.UtbAr_InvAr |  -.0483572   .0235924    -2.05   0.041    -.0947947   -.0019197
                           |
c.StodNyD#c.Barn0_15_InvAr |   .0406762   .0419569     0.97   0.333    -.0419085    .1232608
--------------------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             19.031
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.0e+04
                         (Kleibergen-Paap rk Wald F statistic):         21.670
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen intT
Included instruments: Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
                      c.StodNyD#c.Kvinna c.StodNyD#c.Gift_InvAr
                      c.StodNyD#c.UtbAr_InvAr c.StodNyD#c.Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr intI
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------------+
               Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------------+-------------------------------------------------|
                  InKommun |            0             286            286 *   | 
        InKommun#c.StodNyD |          284             286              2 ?   | 
                     InvAr |            4               5              1     | 
           InvAr#c.StodNyD |            5               5              0 ?   | 
                   IFAUkod |           15              16              1     | 
         IFAUkod#c.StodNyD |           16              16              0 ?   | 
                        Ar |            6               7              1 ?   | 
              Ar#c.StodNyD |            7               7              0 ?   | 
                 FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.StodNyD |           42              42              0 ?   | 
-----------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo ca1

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    329,051    .5449003    7.361609          0        100

.   
.   lincom _b[intT]

 ( 1)  intT = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |   .2309485   .2870416     0.80   0.422    -.3340421     .795939
------------------------------------------------------------------------------

.   lincom _b[logpEnklavVuxen]

 ( 1)  logpEnklavVuxen = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |  -.4438094   .2180564    -2.04   0.043    -.8730147   -.0146041
------------------------------------------------------------------------------

.   lincom _b[logpEnklavVuxen]+_b[intT]

 ( 1)  logpEnklavVuxen + intT = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |   -.212861   .2158434    -0.99   0.325    -.6377104    .2119884
------------------------------------------------------------------------------

.   
.   esttab cb1 cb2 cb3 cb4 using "`output'\Output\TableA7_B.rtf", keep(`controls2' logpEnklavVuxen) ///
>                  replace star(* .1 ** .05 *** .01) b(3) se(3)
(note: file C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\TableA7_B.rtf not found)
(output written to C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\TableA7_B.rtf)

.   
.  //########################## Divide the results by the disproportionality #############################//
.  
.   local sample "inrange(InvAr, 1987, 1991) & inrange(InvAlder, 18, 55) & inrange(Ar, 1991, 2014) & FlyktInv == 1 & Rostratt == 1 "

.   local controls1 "InvAr IFAUkod Ar FodelseAr"

.   local controls2 "Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr"

.   local controls3 "LoneInk UtbAr Sysselsatt" 

.   
.   sum dispropo if `sample', detail

                (firstnm) disproportionality
-------------------------------------------------------------
      Percentiles      Smallest
 1%     .9568092       .5089065
 5%     1.256824       .5089065
10%     1.437346       .5089065       Obs             349,383
25%     1.907957       .5089065       Sum of Wgt.     349,383

50%     2.773606                      Mean           2.909533
                        Largest       Std. Dev.      1.197777
75%     4.018739       6.125851
90%     4.694489       6.125851       Variance        1.43467
95%     4.786949       6.125851       Skewness       .2745827
99%     5.043346       6.125851       Kurtosis       1.877486

.   gen Disp = disprop > r(p50) if disprop <.

.  
.   reghdfe Nominerad (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample' & Disp == 0, absorb(InKommun `controls1') cluster(InKommun) stages
> (first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    174,792
Absorbing 5 HDFE groups                           F(   1,    285) =     201.78
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4278
                                                  Adj R-squared   =     0.4267
                                                  Within R-sq.    =     0.0760
Number of clusters (InKommun) =        286        Root MSE        =     0.7719

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .3215386   .0226356    14.21   0.000     .2769845    .3660927
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   174792
                                                      F(  1,   285) =     6.93
                                                      Prob > F      =   0.0089
Total (centered) SS     =   14888931.3                Centered R2   =   0.0108
Total (uncentered) SS   =   14888931.3                Uncentered R2 =        .
Residual SS             =  14875829.67                Root MSE      =    9.235

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.5603409   .2128888    -2.63   0.009    -.9793746   -.1413071
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             47.976
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.4e+04
                         (Kleibergen-Paap rk Wald F statistic):        201.783
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             286            286 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo cc1

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    174,792    .8678887    9.275566          0        100

.   
.   reghdfe Nominerad (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample' & Disp == 1, absorb(InKommun `controls1') cluster(InKommun) stages
> (first) old 
(running historical version of reghdfe)
(dropped 2 singleton observations)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    174,589
Absorbing 5 HDFE groups                           F(   1,    283) =      32.03
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.5277
                                                  Adj R-squared   =     0.5268
                                                  Within R-sq.    =     0.0533
Number of clusters (InKommun) =        284        Root MSE        =     0.4918

                                     (Std. Err. adjusted for 284 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .1920162    .033929     5.66   0.000      .125231    .2588014
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             284            284 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    284                Number of obs =   174589
                                                      F(  1,   283) =     1.08
                                                      Prob > F      =   0.2990
Total (centered) SS     =  5156738.197                Centered R2   =   0.0058
Total (uncentered) SS   =  5156738.197                Uncentered R2 =        .
Residual SS             =  5154595.109                Root MSE      =    5.439

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.1779353   .1709973    -1.04   0.299    -.5145233    .1586527
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             15.577
                                                   Chi-sq(1) P-val =    0.0001
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):             9829.713
                         (Kleibergen-Paap rk Wald F statistic):         32.028
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             284            284 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo cc2

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    174,589    .2978424    5.449376          0        100

.   
.   reghdfe Nominerad `controls2' (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample' & Disp == 0, absorb(InKommun `controls1') cluster(InKo
> mmun) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    174,792
Absorbing 5 HDFE groups                           F(   5,    285) =      74.53
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4297
                                                  Adj R-squared   =     0.4285
                                                  Within R-sq.    =     0.0790
Number of clusters (InKommun) =        286        Root MSE        =     0.7706

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .3177406   .0226123    14.05   0.000     .2732322     .362249
              Kvinna |  -.0006574    .006759    -0.10   0.923    -.0139613    .0126464
          Gift_InvAr |   .0293905   .0115982     2.53   0.012     .0065615    .0522196
         UtbAr_InvAr |  -.0128443   .0015966    -8.04   0.000    -.0159869   -.0097017
      Barn0_15_InvAr |   .0168585    .003935     4.28   0.000     .0091132    .0246038
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   174792
                                                      F(  5,   285) =    26.44
                                                      Prob > F      =   0.0000
Total (centered) SS     =   14888931.3                Centered R2   =   0.0127
Total (uncentered) SS   =   14888931.3                Uncentered R2 =        .
Residual SS             =   14847358.9                Root MSE      =    9.226

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.4779939   .2128797    -2.25   0.026    -.8970098    -.058978
         Kvinna |  -.2940788   .0631352    -4.66   0.000    -.4183494   -.1698083
     Gift_InvAr |   .3200014   .0930777     3.44   0.001     .1367944    .5032084
    UtbAr_InvAr |   .1344445   .0164918     8.15   0.000     .1019832    .1669057
 Barn0_15_InvAr |  -.0468063   .0335027    -1.40   0.163    -.1127504    .0191378
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             47.792
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.4e+04
                         (Kleibergen-Paap rk Wald F statistic):        197.449
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             286            286 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo cc3

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    174,792    .8678887    9.275566          0        100

.   
.   reghdfe Nominerad `controls2' (logpEnklavVuxen = logpEnklavVuxenInvAr) if `sample' & Disp == 1, absorb(InKommun `controls1') cluster(InKo
> mmun) stages(first) old
(running historical version of reghdfe)
(dropped 2 singleton observations)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    174,589
Absorbing 5 HDFE groups                           F(   5,    283) =       8.55
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.5282
                                                  Adj R-squared   =     0.5273
                                                  Within R-sq.    =     0.0543
Number of clusters (InKommun) =        284        Root MSE        =     0.4916

                                     (Std. Err. adjusted for 284 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .1906948   .0336578     5.67   0.000     .1244435    .2569462
              Kvinna |  -.0103053   .0045287    -2.28   0.024    -.0192195   -.0013911
          Gift_InvAr |   .0082447    .005571     1.48   0.140    -.0027212    .0192106
         UtbAr_InvAr |  -.0055041   .0013541    -4.06   0.000    -.0081695   -.0028387
      Barn0_15_InvAr |   .0043534   .0025996     1.67   0.095    -.0007636    .0094705
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             284            284 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    284                Number of obs =   174589
                                                      F(  5,   283) =     8.62
                                                      Prob > F      =   0.0000
Total (centered) SS     =  5156738.197                Centered R2   =   0.0064
Total (uncentered) SS   =  5156738.197                Uncentered R2 =        .
Residual SS             =  5151406.847                Root MSE      =    5.437

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.1200329   .1720664    -0.70   0.486    -.4587252    .2186594
         Kvinna |  -.0551176   .0326177    -1.69   0.092    -.1193216    .0090864
     Gift_InvAr |   .0199971   .0480934     0.42   0.678    -.0746691    .1146633
    UtbAr_InvAr |   .0566647   .0094458     6.00   0.000     .0380718    .0752575
 Barn0_15_InvAr |   .0196808   .0140476     1.40   0.162    -.0079702    .0473318
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             15.637
                                                   Chi-sq(1) P-val =    0.0001
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):             9675.122
                         (Kleibergen-Paap rk Wald F statistic):         32.100
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             284            284 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo cc4

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    174,589    .2978424    5.449376          0        100

.     
.  // Code to significance the difference, use a fully interacted model to replicate the split sample results 
.   capture drop intT intI

.   gen intT = logpEnklavVuxen*Disp

.   gen intI = logpEnklavVuxenInvAr*Disp

.   
.   reghdfe Nominerad  `controls2' c.Disp#c.Kvinna c.Disp#c.Gift_InvAr c.Disp#c.UtbAr_InvAr c.Disp#c.Barn0_15_InvAr (logpEnklavVuxen intT = l
> ogpEnklavVuxenInvAr intI) if `sample', absorb(InKommun##c.Disp InvAr##c.Disp IFAUkod##c.Disp Ar##c.Disp FodelseAr##c.Disp) cluster(InKommun
> ) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    349,383
Absorbing 5 HDFE groups                           F(  10,    285) =      38.91
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4846
                                                  Adj R-squared   =     0.4836
                                                  Within R-sq.    =     0.0720
Number of clusters (InKommun) =        286        Root MSE        =     0.6464

                                        (Std. Err. adjusted for 286 clusters in InKommun)
-----------------------------------------------------------------------------------------
                        |               Robust
        logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------------+----------------------------------------------------------------
   logpEnklavVuxenInvAr |   .3177406   .0226218    14.05   0.000     .2732136    .3622675
                   intI |  -.1270457   .0358852    -3.54   0.000    -.1976795    -.056412
                 Kvinna |  -.0006574   .0067618    -0.10   0.923    -.0139668     .012652
             Gift_InvAr |   .0293905   .0116031     2.53   0.012     .0065519    .0522291
            UtbAr_InvAr |  -.0128443   .0015972    -8.04   0.000    -.0159882   -.0097004
         Barn0_15_InvAr |   .0168585   .0039366     4.28   0.000       .00911     .024607
                        |
        c.Disp#c.Kvinna |  -.0096478   .0063418    -1.52   0.129    -.0221306    .0028349
                        |
    c.Disp#c.Gift_InvAr |  -.0211458   .0118169    -1.79   0.075    -.0444053    .0021137
                        |
   c.Disp#c.UtbAr_InvAr |   .0073402   .0018483     3.97   0.000     .0037022    .0109783
                        |
c.Disp#c.Barn0_15_InvAr |  -.0125051   .0042278    -2.96   0.003    -.0208267   -.0041835
-----------------------------------------------------------------------------------------

Absorbed degrees of freedom:
--------------------------------------------------------------------------+
            Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
------------------------+-------------------------------------------------|
               InKommun |            0             286            286 *   | 
        InKommun#c.Disp |          286             286              0 ?   | 
                  InvAr |            4               5              1     | 
           InvAr#c.Disp |            5               5              0 ?   | 
                IFAUkod |           15              16              1     | 
         IFAUkod#c.Disp |           16              16              0 ?   | 
                     Ar |            6               7              1 ?   | 
              Ar#c.Disp |            7               7              0 ?   | 
              FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.Disp |           42              42              0 ?   | 
--------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation
WARNING: Missing F statistic (dropped variables due to collinearity or too few clusters).

Stage: first - intT

HDFE Linear regression                            Number of obs   =    349,383
Absorbing 5 HDFE groups                           F(   5,    285) =          .
Statistics robust to heteroskedasticity           Prob > F        =          .
                                                  R-squared       =     0.9768
                                                  Adj R-squared   =     0.9767
                                                  Within R-sq.    =     0.0543
Number of clusters (InKommun) =        286        Root MSE        =     0.3475

                                        (Std. Err. adjusted for 286 clusters in InKommun)
-----------------------------------------------------------------------------------------
                        |               Robust
                   intT |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------------+----------------------------------------------------------------
   logpEnklavVuxenInvAr |  -1.24e-16          .        .       .            .           .
                   intI |   .1906948   .0336714     5.66   0.000     .1244187     .256971
                 Kvinna |   9.36e-17   1.16e-10     0.00   1.000    -2.28e-10    2.28e-10
             Gift_InvAr |  -9.62e-17   1.70e-10    -0.00   1.000    -3.34e-10    3.34e-10
            UtbAr_InvAr |   3.52e-17          .        .       .            .           .
         Barn0_15_InvAr |   2.12e-17          .        .       .            .           .
                        |
        c.Disp#c.Kvinna |  -.0103053   .0045305    -2.27   0.024    -.0192228   -.0013878
                        |
    c.Disp#c.Gift_InvAr |   .0082447   .0055733     1.48   0.140    -.0027253    .0192147
                        |
   c.Disp#c.UtbAr_InvAr |  -.0055041   .0013547    -4.06   0.000    -.0081705   -.0028377
                        |
c.Disp#c.Barn0_15_InvAr |   .0043534   .0026007     1.67   0.095    -.0007656    .0094724
-----------------------------------------------------------------------------------------

Absorbed degrees of freedom:
--------------------------------------------------------------------------+
            Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
------------------------+-------------------------------------------------|
               InKommun |            0             286            286 *   | 
        InKommun#c.Disp |          286             286              0 ?   | 
                  InvAr |            4               5              1     | 
           InvAr#c.Disp |            5               5              0 ?   | 
                IFAUkod |           15              16              1     | 
         IFAUkod#c.Disp |           16              16              0 ?   | 
                     Ar |            6               7              1 ?   | 
              Ar#c.Disp |            7               7              0 ?   | 
              FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.Disp |           42              42              0 ?   | 
--------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   349383
                                                      F( 10,   285) =    14.80
                                                      Prob > F      =   0.0000
Total (centered) SS     =   20045669.5                Centered R2   =   0.0125
Total (uncentered) SS   =   20045669.5                Uncentered R2 =        .
Residual SS             =  19998765.74                Root MSE      =    7.574

-----------------------------------------------------------------------------------------
                        |               Robust
              Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------------+----------------------------------------------------------------
        logpEnklavVuxen |  -.4779939   .2129685    -2.24   0.026    -.8971845   -.0588033
                   intT |    .357961   .2398394     1.49   0.137    -.1141203    .8300423
                 Kvinna |  -.2940788   .0631616    -4.66   0.000    -.4184012   -.1697565
             Gift_InvAr |   .3200014   .0931166     3.44   0.001      .136718    .5032848
            UtbAr_InvAr |   .1344445   .0164987     8.15   0.000     .1019697    .1669192
         Barn0_15_InvAr |  -.0468063   .0335167    -1.40   0.164    -.1127779    .0191653
                        |
        c.Disp#c.Kvinna |   .2389612   .0645846     3.70   0.000     .1118379    .3660846
                        |
    c.Disp#c.Gift_InvAr |  -.3000043   .0973986    -3.08   0.002    -.4917161   -.1082925
                        |
   c.Disp#c.UtbAr_InvAr |  -.0777798   .0157172    -4.95   0.000    -.1087164   -.0468432
                        |
c.Disp#c.Barn0_15_InvAr |   .0664871   .0328069     2.03   0.044     .0019126    .1310616
-----------------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             15.637
                                                   Chi-sq(1) P-val =    0.0001
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):             9672.766
                         (Kleibergen-Paap rk Wald F statistic):         16.037
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen intT
Included instruments: Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
                      c.Disp#c.Kvinna c.Disp#c.Gift_InvAr c.Disp#c.UtbAr_InvAr
                      c.Disp#c.Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr intI
------------------------------------------------------------------------------

Absorbed degrees of freedom:
--------------------------------------------------------------------------+
            Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
------------------------+-------------------------------------------------|
               InKommun |            0             286            286 *   | 
        InKommun#c.Disp |          286             286              0 ?   | 
                  InvAr |            4               5              1     | 
           InvAr#c.Disp |            5               5              0 ?   | 
                IFAUkod |           15              16              1     | 
         IFAUkod#c.Disp |           16              16              0 ?   | 
                     Ar |            6               7              1 ?   | 
              Ar#c.Disp |            7               7              0 ?   | 
              FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.Disp |           42              42              0 ?   | 
--------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo ca1

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    349,383    .5830278    7.613345          0        100

.   
.   lincom _b[intT]

 ( 1)  intT = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |    .357961   .2398394     1.49   0.137    -.1141203    .8300423
------------------------------------------------------------------------------

.   lincom _b[logpEnklavVuxen]

 ( 1)  logpEnklavVuxen = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |  -.4779939   .2129685    -2.24   0.026    -.8971845   -.0588033
------------------------------------------------------------------------------

.   lincom _b[logpEnklavVuxen]+_b[intT]

 ( 1)  logpEnklavVuxen + intT = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |  -.1200329    .172136    -0.70   0.486     -.458852    .2187862
------------------------------------------------------------------------------

.   
.   esttab cc1 cc2 cc3 cc4 using "`output'\Output\TableA7_C.rtf", keep(`controls2' logpEnklavVuxen) ///
>                  replace star(* .1 ** .05 *** .01) b(3) se(3)
(note: file C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\TableA7_C.rtf not found)
(output written to C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\TableA7_C.rtf)

.   
.   
.   est restore ca3
(results ca3 are active now)

.   gen b = _b[logpEnklavVuxen] in 1
(349,382 missing values generated)

.   gen lci = b-1.96*_se[logpEnklavVuxen] in 1 
(349,382 missing values generated)

.   gen hci = b+1.96*_se[logpEnklavVuxen] in 1
(349,382 missing values generated)

.   
.   est restore ca4
(results ca4 are active now)

.   replace b = _b[logpEnklavVuxen] in 2
(1 real change made)

.   replace lci = b-1.96*_se[logpEnklavVuxen] in 2 
(1 real change made)

.   replace hci = b+1.96*_se[logpEnklavVuxen] in 2
(1 real change made)

.   
.   est restore cb3
(results cb3 are active now)

.   replace b = _b[logpEnklavVuxen] in 3
(1 real change made)

.   replace lci = b-1.96*_se[logpEnklavVuxen] in 3 
(1 real change made)

.   replace hci = b+1.96*_se[logpEnklavVuxen] in 3
(1 real change made)

.   
.   est restore cb4
(results cb4 are active now)

.   replace b = _b[logpEnklavVuxen] in 4
(1 real change made)

.   replace lci = b-1.96*_se[logpEnklavVuxen] in 4 
(1 real change made)

.   replace hci = b+1.96*_se[logpEnklavVuxen] in 4
(1 real change made)

.   
.   est restore cc3
(results cc3 are active now)

.   replace b = _b[logpEnklavVuxen] in 5
(1 real change made)

.   replace lci = b-1.96*_se[logpEnklavVuxen] in 5 
(1 real change made)

.   replace hci = b+1.96*_se[logpEnklavVuxen] in 5
(1 real change made)

.   
.   est restore cc4
(results cc4 are active now)

.   replace b = _b[logpEnklavVuxen] in 6
(1 real change made)

.   replace lci = b-1.96*_se[logpEnklavVuxen] in 6 
(1 real change made)

.   replace hci = b+1.96*_se[logpEnklavVuxen] in 6
(1 real change made)

.   
.   gen hid = .
(349,383 missing values generated)

.   replace hid = 1 in 1
(1 real change made)

.   replace hid = 1.5 in 2
(1 real change made)

.   replace hid = 2.5 in 3
(1 real change made)

.   replace hid = 3 in 4
(1 real change made)

.   replace hid = 4 in 5
(1 real change made)

.   replace hid = 4.5 in 6
(1 real change made)

.   
.   twoway (rcap lci hci hid if inrange(hid, 1, 4.5), lcol(black)) (scatter b hid if inrange(hid, 1, 4.5), legend(off) xtitle("") mcol(black)
>  ///
>           xscale(range(.5 5)) xlab(1 "    Low" 1.5 "    High" 2.5 "    Low" 3 "    High" 4 "    Low" 4.5 "    High", angle(45) labsize(smal
> l)) ///
>                   text(-1.25 1.25 "Left party strength", place(c) color(black)) text(-1.25 2.75 "Right populist strength", place(c) color(b
> lack)) ///
>                   text(-1.25 4.25 "Disproportionality", place(c) color(black)) graphregion(color(white)) ylab(-1(.25).5))

.   
.   graph export "C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\Figure3.png", as(png) replace    
(note: file C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\Figure3.png not found)
(file C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\Figure3.png written in PNG format)

.   
.  //################# Results by sex  ###############################/
.    local sample "inrange(InvAr, 1987, 1991) & inrange(InvAlder, 18, 55) & inrange(Ar, 1991, 2014) & FlyktInv == 1 & Rostratt == 1 "

.   local controls1 "InvAr IFAUkod Ar FodelseAr"

.   local controls2 "Gift_InvAr UtbAr_InvAr Barn0_15_InvAr"

.   local controls3 "LoneInk UtbAr Sysselsatt" 

.  
.   reghdfe Nominerad (logpEnklavVuxen = logpEnklavVuxenInvAr) if Kvinna == 0 & `sample', absorb(InKommun `controls1') cluster(InKommun) stag
> es(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    203,217
Absorbing 5 HDFE groups                           F(   1,    283) =     130.42
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.3669
                                                  Adj R-squared   =     0.3658
                                                  Within R-sq.    =     0.0473
Number of clusters (InKommun) =        284        Root MSE        =     0.7178

                                     (Std. Err. adjusted for 284 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .2447717    .021433    11.42   0.000     .2025834    .2869599
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             284            284 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    284                Number of obs =   203217
                                                      F(  1,   283) =     3.46
                                                      Prob > F      =   0.0638
Total (centered) SS     =  13420775.48                Centered R2   =   0.0096
Total (uncentered) SS   =  13420775.48                Uncentered R2 =        .
Residual SS             =  13398510.57                Root MSE      =    8.127

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.3852933   .2070545    -1.86   0.064    -.7928557    .0222691
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             43.616
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.0e+04
                         (Kleibergen-Paap rk Wald F statistic):        130.424
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             284            284 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo hs1

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    203,217    .6702195    8.159233          0        100

.   
.   reghdfe Nominerad (logpEnklavVuxen = logpEnklavVuxenInvAr) if Kvinna == 1 & `sample', absorb(InKommun `controls1') cluster(InKommun) stag
> es(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    146,166
Absorbing 5 HDFE groups                           F(   1,    285) =     167.59
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4725
                                                  Adj R-squared   =     0.4712
                                                  Within R-sq.    =     0.0885
Number of clusters (InKommun) =        286        Root MSE        =     0.6519

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |    .311701   .0240774    12.95   0.000      .264309     .359093
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   146166
                                                      F(  1,   285) =     5.64
                                                      Prob > F      =   0.0183
Total (centered) SS     =  6648563.388                Centered R2   =   0.0115
Total (uncentered) SS   =  6648563.388                Uncentered R2 =        .
Residual SS             =  6641259.512                Root MSE      =    6.749

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |   -.448859     .18907    -2.37   0.018    -.8210098   -.0767082
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             39.727
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.4e+04
                         (Kleibergen-Paap rk Wald F statistic):        167.594
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             286            286 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo hs2

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    146,166    .4618037    6.779928          0        100

.   
.   reghdfe Nominerad `controls2' (logpEnklavVuxen = logpEnklavVuxenInvAr) if Kvinna == 0 & `sample', absorb(InKommun `controls1') cluster(In
> Kommun) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    203,217
Absorbing 5 HDFE groups                           F(   4,    283) =      44.96
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.3677
                                                  Adj R-squared   =     0.3666
                                                  Within R-sq.    =     0.0485
Number of clusters (InKommun) =        284        Root MSE        =     0.7173

                                     (Std. Err. adjusted for 284 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .2423065    .021268    11.39   0.000     .2004428    .2841701
          Gift_InvAr |   .0082913   .0096488     0.86   0.391    -.0107012    .0272839
         UtbAr_InvAr |  -.0094868   .0013636    -6.96   0.000    -.0121709   -.0068026
      Barn0_15_InvAr |   .0064782   .0033605     1.93   0.055    -.0001364    .0130929
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             284            284 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    284                Number of obs =   203217
                                                      F(  4,   283) =    14.46
                                                      Prob > F      =   0.0000
Total (centered) SS     =  13420775.48                Centered R2   =   0.0106
Total (uncentered) SS   =  13420775.48                Uncentered R2 =        .
Residual SS             =  13385837.81                Root MSE      =    8.123

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |   -.309931   .2107556    -1.47   0.143    -.7247786    .1049165
     Gift_InvAr |   .2838754   .0795103     3.57   0.000     .1273687    .4403821
    UtbAr_InvAr |   .0996661   .0162148     6.15   0.000     .0677491     .131583
 Barn0_15_InvAr |     -.0046   .0270001    -0.17   0.865    -.0577466    .0485466
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             43.655
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):             9847.251
                         (Kleibergen-Paap rk Wald F statistic):        129.800
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             284            284 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo hs3

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    203,217    .6702195    8.159233          0        100

.   
.   reghdfe Nominerad `controls2' (logpEnklavVuxen = logpEnklavVuxenInvAr) if Kvinna == 1 & `sample', absorb(InKommun `controls1') cluster(In
> Kommun) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    146,166
Absorbing 5 HDFE groups                           F(   4,    285) =      43.36
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4734
                                                  Adj R-squared   =     0.4721
                                                  Within R-sq.    =     0.0900
Number of clusters (InKommun) =        286        Root MSE        =     0.6513

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .3098534   .0237824    13.03   0.000     .2630419    .3566649
          Gift_InvAr |   .0102391   .0088337     1.16   0.247    -.0071484    .0276266
         UtbAr_InvAr |  -.0085523   .0018453    -4.63   0.000    -.0121844   -.0049201
      Barn0_15_InvAr |   .0105567   .0039889     2.65   0.009     .0027053     .018408
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   146166
                                                      F(  4,   285) =    11.47
                                                      Prob > F      =   0.0000
Total (centered) SS     =  6648563.388                Centered R2   =   0.0126
Total (uncentered) SS   =  6648563.388                Uncentered R2 =        .
Residual SS             =  6633871.035                Root MSE      =    6.745

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.3939498   .1901845    -2.07   0.039    -.7682943   -.0196054
     Gift_InvAr |   .0468064   .0701138     0.67   0.505    -.0912001    .1848129
    UtbAr_InvAr |   .0903583   .0145721     6.20   0.000     .0616756     .119041
 Barn0_15_InvAr |   .0137348   .0277235     0.50   0.621    -.0408341    .0683037
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             40.113
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.4e+04
                         (Kleibergen-Paap rk Wald F statistic):        169.746
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             286            286 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo hs4

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    146,166    .4618037    6.779928          0        100

.   
.  // Code to significance the difference, use a fully interacted model to replicate the split sample results 
.   capture drop intT intI

.   gen intT = logpEnklavVuxen*Kvinna

.   gen intI = logpEnklavVuxenInvAr*Kvinna

.   
.   reghdfe Nominerad  `controls2' c.Kvinna#c.Gift_InvAr c.Kvinna#c.UtbAr_InvAr c.Kvinna#c.Barn0_15_InvAr (logpEnklavVuxen intT = logpEnklavV
> uxenInvAr intI) if `sample', absorb(InKommun##c.Kvinna InvAr##c.Kvinna IFAUkod##c.Kvinna Ar##c.Kvinna FodelseAr##c.Kvinna) cluster(InKommun
> ) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    349,383
Absorbing 5 HDFE groups                           F(   8,    285) =      27.36
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4120
                                                  Adj R-squared   =     0.4108
                                                  Within R-sq.    =     0.0644
Number of clusters (InKommun) =        286        Root MSE        =     0.6905

                                          (Std. Err. adjusted for 286 clusters in InKommun)
-------------------------------------------------------------------------------------------
                          |               Robust
          logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
--------------------------+----------------------------------------------------------------
     logpEnklavVuxenInvAr |   .2423065   .0212772    11.39   0.000     .2004261    .2841868
                     intI |   .0675469    .008676     7.79   0.000     .0504698    .0846241
               Gift_InvAr |   .0082913    .009653     0.86   0.391    -.0107088    .0272915
              UtbAr_InvAr |  -.0094868   .0013642    -6.95   0.000    -.0121719   -.0068016
           Barn0_15_InvAr |   .0064782   .0033619     1.93   0.055    -.0001391    .0130955
                          |
    c.Kvinna#c.Gift_InvAr |   .0019477   .0111887     0.17   0.862    -.0200753    .0239708
                          |
   c.Kvinna#c.UtbAr_InvAr |   .0009345   .0020597     0.45   0.650    -.0031197    .0049887
                          |
c.Kvinna#c.Barn0_15_InvAr |   .0040784   .0045374     0.90   0.369    -.0048526    .0130095
-------------------------------------------------------------------------------------------

Absorbed degrees of freedom:
----------------------------------------------------------------------------+
              Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
--------------------------+-------------------------------------------------|
                 InKommun |            0             286            286 *   | 
        InKommun#c.Kvinna |          284             286              2 ?   | 
                    InvAr |            4               5              1     | 
           InvAr#c.Kvinna |            5               5              0 ?   | 
                  IFAUkod |           15              16              1     | 
         IFAUkod#c.Kvinna |           16              16              0 ?   | 
                       Ar |            6               7              1 ?   | 
              Ar#c.Kvinna |            7               7              0 ?   | 
                FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.Kvinna |           42              42              0 ?   | 
----------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation
WARNING: Missing F statistic (dropped variables due to collinearity or too few clusters).

Stage: first - intT

HDFE Linear regression                            Number of obs   =    349,383
Absorbing 5 HDFE groups                           F(   5,    285) =          .
Statistics robust to heteroskedasticity           Prob > F        =          .
                                                  R-squared       =     0.9684
                                                  Adj R-squared   =     0.9683
                                                  Within R-sq.    =     0.0900
Number of clusters (InKommun) =        286        Root MSE        =     0.4212

                                          (Std. Err. adjusted for 286 clusters in InKommun)
-------------------------------------------------------------------------------------------
                          |               Robust
                     intT |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
--------------------------+----------------------------------------------------------------
     logpEnklavVuxenInvAr |  -4.72e-16   6.13e-10    -0.00   1.000    -1.21e-09    1.21e-09
                     intI |   .3098534   .0237913    13.02   0.000     .2630243    .3566824
               Gift_InvAr |  -1.14e-16   1.54e-10    -0.00   1.000    -3.03e-10    3.03e-10
              UtbAr_InvAr |  -3.24e-17   4.58e-11    -0.00   1.000    -9.02e-11    9.02e-11
           Barn0_15_InvAr |  -8.01e-17          .        .       .            .           .
                          |
    c.Kvinna#c.Gift_InvAr |   .0102391    .008837     1.16   0.248     -.007155    .0276331
                          |
   c.Kvinna#c.UtbAr_InvAr |  -.0085523    .001846    -4.63   0.000    -.0121857   -.0049188
                          |
c.Kvinna#c.Barn0_15_InvAr |   .0105567   .0039904     2.65   0.009     .0027024     .018411
-------------------------------------------------------------------------------------------

Absorbed degrees of freedom:
----------------------------------------------------------------------------+
              Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
--------------------------+-------------------------------------------------|
                 InKommun |            0             286            286 *   | 
        InKommun#c.Kvinna |          284             286              2 ?   | 
                    InvAr |            4               5              1     | 
           InvAr#c.Kvinna |            5               5              0 ?   | 
                  IFAUkod |           15              16              1     | 
         IFAUkod#c.Kvinna |           16              16              0 ?   | 
                       Ar |            6               7              1 ?   | 
              Ar#c.Kvinna |            7               7              0 ?   | 
                FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.Kvinna |           42              42              0 ?   | 
----------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   349383
                                                      F(  8,   285) =    10.67
                                                      Prob > F      =   0.0000
Total (centered) SS     =  20069338.87                Centered R2   =   0.0114
Total (uncentered) SS   =  20069338.87                Uncentered R2 =        .
Residual SS             =  20019708.85                Root MSE      =    7.577

-------------------------------------------------------------------------------------------
                          |               Robust
                Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
--------------------------+----------------------------------------------------------------
          logpEnklavVuxen |   -.309931   .2108461    -1.47   0.143    -.7249442    .1050821
                     intT |  -.0840188   .2355634    -0.36   0.722    -.5476836    .3796461
               Gift_InvAr |   .2838754   .0795445     3.57   0.000     .1273063    .4404446
              UtbAr_InvAr |   .0996661   .0162218     6.14   0.000     .0677364    .1315958
           Barn0_15_InvAr |     -.0046   .0270117    -0.17   0.865    -.0577678    .0485678
                          |
    c.Kvinna#c.Gift_InvAr |   -.237069   .0994523    -2.38   0.018    -.4328232   -.0413148
                          |
   c.Kvinna#c.UtbAr_InvAr |  -.0093078   .0195993    -0.47   0.635    -.0478856    .0292701
                          |
c.Kvinna#c.Barn0_15_InvAr |   .0183348     .03524     0.52   0.603    -.0510289    .0876986
-------------------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             43.655
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):             8457.539
                         (Kleibergen-Paap rk Wald F statistic):         64.844
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen intT
Included instruments: Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
                      c.Kvinna#c.Gift_InvAr c.Kvinna#c.UtbAr_InvAr
                      c.Kvinna#c.Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr intI
------------------------------------------------------------------------------

Absorbed degrees of freedom:
----------------------------------------------------------------------------+
              Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
--------------------------+-------------------------------------------------|
                 InKommun |            0             286            286 *   | 
        InKommun#c.Kvinna |          284             286              2 ?   | 
                    InvAr |            4               5              1     | 
           InvAr#c.Kvinna |            5               5              0 ?   | 
                  IFAUkod |           15              16              1     | 
         IFAUkod#c.Kvinna |           16              16              0 ?   | 
                       Ar |            6               7              1 ?   | 
              Ar#c.Kvinna |            7               7              0 ?   | 
                FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.Kvinna |           42              42              0 ?   | 
----------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo ca1

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    349,383    .5830278    7.613345          0        100

.   
.   lincom _b[intT]

 ( 1)  intT = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |  -.0840188   .2355634    -0.36   0.722    -.5476836    .3796461
------------------------------------------------------------------------------

.   lincom _b[logpEnklavVuxen]

 ( 1)  logpEnklavVuxen = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |   -.309931   .2108461    -1.47   0.143    -.7249442    .1050821
------------------------------------------------------------------------------

.   lincom _b[logpEnklavVuxen]+_b[intT]

 ( 1)  logpEnklavVuxen + intT = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |  -.3939498   .1902559    -2.07   0.039    -.7684348   -.0194649
------------------------------------------------------------------------------

.   
.   esttab hs1 hs2 hs3 hs4 using "`output'\Output\TableA8_A.rtf", keep(`controls2' logpEnklavVuxen) replace star(* .1 ** .05 *** .01) b(3) se
> (3)
(note: file C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\TableA8_A.rtf not found)
(output written to C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\TableA8_A.rtf)

.  
.  //################# Results by education  ###############################/
.   local sample "inrange(InvAr, 1987, 1991) & inrange(InvAlder, 18, 55) & inrange(Ar, 1991, 2014) & FlyktInv == 1 & Rostratt == 1 "

.   local controls1 "InvAr IFAUkod Ar FodelseAr"

.   local controls2 "Kvinna Gift_InvAr Barn0_15_InvAr"

.   local controls3 "LoneInk UtbAr Sysselsatt" 

.   
.   gen HUtb = UtbAr_InvAr > 10 if UtbAr_InvAr <.

.   
.   reghdfe Nominerad (logpEnklavVuxen = logpEnklavVuxenInvAr) if HUtb == 0 & `sample', absorb(InKommun `controls1') cluster(InKommun) stages
> (first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    143,544
Absorbing 5 HDFE groups                           F(   1,    284) =     136.01
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4394
                                                  Adj R-squared   =     0.4381
                                                  Within R-sq.    =     0.0632
Number of clusters (InKommun) =        285        Root MSE        =     0.6892

                                     (Std. Err. adjusted for 285 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .2775631   .0237998    11.66   0.000     .2307167    .3244096
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             285            285 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    285                Number of obs =   143544
                                                      F(  1,   284) =     3.23
                                                      Prob > F      =   0.0734
Total (centered) SS     =  4824050.392                Centered R2   =   0.0105
Total (uncentered) SS   =  4824050.392                Uncentered R2 =        .
Residual SS             =   4821999.19                Root MSE      =    5.803

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.3348587   .1863618    -1.80   0.073    -.7016843    .0319669
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             40.291
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):             9680.264
                         (Kleibergen-Paap rk Wald F statistic):        136.012
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             285            285 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo hu1

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    143,544    .3406621    5.826697          0        100

.   
.   reghdfe Nominerad (logpEnklavVuxen = logpEnklavVuxenInvAr) if HUtb == 1 & `sample', absorb(InKommun `controls1') cluster(InKommun) stages
> (first) old
(running historical version of reghdfe)
(dropped 1 singleton observations)
(converged in 9 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    205,838
Absorbing 5 HDFE groups                           F(   1,    284) =     144.39
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.3939
                                                  Adj R-squared   =     0.3928
                                                  Within R-sq.    =     0.0569
Number of clusters (InKommun) =        285        Root MSE        =     0.6894

                                     (Std. Err. adjusted for 285 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .2623199   .0218307    12.02   0.000     .2193494    .3052903
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             285            285 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    285                Number of obs =   205838
                                                      F(  1,   284) =     3.55
                                                      Prob > F      =   0.0607
Total (centered) SS     =  15203717.46                Centered R2   =   0.0122
Total (uncentered) SS   =  15203717.46                Uncentered R2 =        .
Residual SS             =   15176075.5                Root MSE      =    8.594

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.4447665   .2362047    -1.88   0.061    -.9097005    .0201675
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             42.879
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.2e+04
                         (Kleibergen-Paap rk Wald F statistic):        144.387
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             285            285 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo hu2

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    205,838    .7520477    8.639419          0        100

.   
.   reghdfe Nominerad `controls2' (logpEnklavVuxen = logpEnklavVuxenInvAr) if HUtb == 0 & `sample', absorb(InKommun `controls1') cluster(InKo
> mmun) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    143,544
Absorbing 5 HDFE groups                           F(   4,    284) =      44.49
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4399
                                                  Adj R-squared   =     0.4385
                                                  Within R-sq.    =     0.0640
Number of clusters (InKommun) =        285        Root MSE        =     0.6889

                                     (Std. Err. adjusted for 285 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .2769211   .0237147    11.68   0.000     .2302422    .3235999
              Kvinna |  -.0155008   .0079245    -1.96   0.051    -.0310989    .0000973
          Gift_InvAr |   .0308121   .0109406     2.82   0.005     .0092771     .052347
      Barn0_15_InvAr |   .0087412   .0037972     2.30   0.022      .001267    .0162153
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             285            285 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    285                Number of obs =   143544
                                                      F(  4,   284) =     3.12
                                                      Prob > F      =   0.0157
Total (centered) SS     =  4824050.392                Centered R2   =   0.0107
Total (uncentered) SS   =  4824050.392                Uncentered R2 =        .
Residual SS             =  4821091.493                Root MSE      =    5.803

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.3457359   .1870983    -1.85   0.066    -.7140112    .0225395
         Kvinna |  -.1741916   .0534252    -3.26   0.001    -.2793512   -.0690321
     Gift_InvAr |   .1030048   .0656765     1.57   0.118    -.0262697    .2322792
 Barn0_15_InvAr |  -.0161521   .0201609    -0.80   0.424    -.0558359    .0235318
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             40.335
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):             9632.898
                         (Kleibergen-Paap rk Wald F statistic):        136.357
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: Kvinna Gift_InvAr Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             285            285 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo hu3

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    143,544    .3406621    5.826697          0        100

.   
.   reghdfe Nominerad `controls2' (logpEnklavVuxen = logpEnklavVuxenInvAr) if HUtb == 1 & `sample', absorb(InKommun `controls1') cluster(InKo
> mmun) stages(first) old
(running historical version of reghdfe)
(dropped 1 singleton observations)
(converged in 9 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    205,838
Absorbing 5 HDFE groups                           F(   4,    284) =      36.28
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.3939
                                                  Adj R-squared   =     0.3929
                                                  Within R-sq.    =     0.0569
Number of clusters (InKommun) =        285        Root MSE        =     0.6894

                                     (Std. Err. adjusted for 285 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .2621336   .0217988    12.03   0.000     .2192259    .3050413
              Kvinna |  -.0076741   .0066483    -1.15   0.249    -.0207603    .0054121
          Gift_InvAr |   .0095899   .0092584     1.04   0.301    -.0086338    .0278136
      Barn0_15_InvAr |   .0025557   .0040258     0.63   0.526    -.0053685    .0104799
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             285            285 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    285                Number of obs =   205838
                                                      F(  4,   284) =     3.33
                                                      Prob > F      =   0.0109
Total (centered) SS     =  15203717.46                Centered R2   =   0.0124
Total (uncentered) SS   =  15203717.46                Uncentered R2 =        .
Residual SS             =  15173510.77                Root MSE      =    8.593

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.4575452   .2364704    -1.93   0.054    -.9230022    .0079118
         Kvinna |  -.1671832   .0594779    -2.81   0.005    -.2842567   -.0501096
     Gift_InvAr |   .1850984   .0870977     2.13   0.034     .0136595    .3565374
 Barn0_15_InvAr |   .0230546   .0347229     0.66   0.507    -.0452924    .0914015
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             42.900
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.2e+04
                         (Kleibergen-Paap rk Wald F statistic):        144.604
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: Kvinna Gift_InvAr Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             285            285 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo hu4

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    205,838    .7520477    8.639419          0        100

.   
.   // Code to significance the difference, use a fully interacted model to replicate the split sample results 
.   capture drop intT intI

.   gen intT = logpEnklavVuxen*HUtb

.   gen intI = logpEnklavVuxenInvAr*HUtb

.   
.   reghdfe Nominerad  `controls2' c.HUtb#c.Gift_InvAr c.Kvinna#c.HUtb c.HUtb#c.Barn0_15_InvAr (logpEnklavVuxen intT = logpEnklavVuxenInvAr i
> ntI) if `sample', absorb(InKommun##c.HUtb InvAr##c.HUtb IFAUkod##c.HUtb Ar##c.HUtb FodelseAr##c.HUtb) cluster(InKommun) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    349,383
Absorbing 5 HDFE groups                           F(   8,    285) =      24.24
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4141
                                                  Adj R-squared   =     0.4129
                                                  Within R-sq.    =     0.0599
Number of clusters (InKommun) =        286        Root MSE        =     0.6892

                                        (Std. Err. adjusted for 286 clusters in InKommun)
-----------------------------------------------------------------------------------------
                        |               Robust
        logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------------+----------------------------------------------------------------
   logpEnklavVuxenInvAr |   .2769211   .0237234    11.67   0.000     .2302258    .3236163
                   intI |  -.0147874    .014247    -1.04   0.300    -.0428301    .0132553
                 Kvinna |  -.0155008   .0079274    -1.96   0.052    -.0311044    .0001028
             Gift_InvAr |   .0308121   .0109446     2.82   0.005     .0092695    .0523546
         Barn0_15_InvAr |   .0087412   .0037986     2.30   0.022     .0012644     .016218
                        |
    c.HUtb#c.Gift_InvAr |  -.0212222   .0137268    -1.55   0.123     -.048241    .0057966
                        |
        c.Kvinna#c.HUtb |   .0078267   .0098797     0.79   0.429    -.0116198    .0272732
                        |
c.HUtb#c.Barn0_15_InvAr |  -.0061855   .0052196    -1.19   0.237    -.0164594    .0040884
-----------------------------------------------------------------------------------------

Absorbed degrees of freedom:
--------------------------------------------------------------------------+
            Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
------------------------+-------------------------------------------------|
               InKommun |            0             286            286 *   | 
        InKommun#c.HUtb |          285             286              1 ?   | 
                  InvAr |            4               5              1     | 
           InvAr#c.HUtb |            5               5              0 ?   | 
                IFAUkod |           15              16              1     | 
         IFAUkod#c.HUtb |           16              16              0 ?   | 
                     Ar |            6               7              1 ?   | 
              Ar#c.HUtb |            7               7              0 ?   | 
              FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.HUtb |           42              42              0 ?   | 
--------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation
WARNING: Missing F statistic (dropped variables due to collinearity or too few clusters).

Stage: first - intT

HDFE Linear regression                            Number of obs   =    349,383
Absorbing 5 HDFE groups                           F(   5,    285) =          .
Statistics robust to heteroskedasticity           Prob > F        =          .
                                                  R-squared       =     0.9508
                                                  Adj R-squared   =     0.9507
                                                  Within R-sq.    =     0.0569
Number of clusters (InKommun) =        286        Root MSE        =     0.5293

                                        (Std. Err. adjusted for 286 clusters in InKommun)
-----------------------------------------------------------------------------------------
                        |               Robust
                   intT |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------------+----------------------------------------------------------------
   logpEnklavVuxenInvAr |   3.02e-15          .        .       .            .           .
                   intI |   .2621336   .0218084    12.02   0.000     .2192077    .3050595
                 Kvinna |  -4.51e-17   3.25e-10    -0.00   1.000    -6.39e-10    6.39e-10
             Gift_InvAr |  -4.91e-17          .        .       .            .           .
         Barn0_15_InvAr |   3.33e-17   7.32e-11     0.00   1.000    -1.44e-10    1.44e-10
                        |
    c.HUtb#c.Gift_InvAr |   .0095899   .0092624     1.04   0.301    -.0086416    .0278213
                        |
        c.Kvinna#c.HUtb |  -.0076741   .0066512    -1.15   0.250    -.0207659    .0054177
                        |
c.HUtb#c.Barn0_15_InvAr |   .0025557   .0040276     0.63   0.526    -.0053719    .0104833
-----------------------------------------------------------------------------------------

Absorbed degrees of freedom:
--------------------------------------------------------------------------+
            Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
------------------------+-------------------------------------------------|
               InKommun |            0             286            286 *   | 
        InKommun#c.HUtb |          285             286              1 ?   | 
                  InvAr |            4               5              1     | 
           InvAr#c.HUtb |            5               5              0 ?   | 
                IFAUkod |           15              16              1     | 
         IFAUkod#c.HUtb |           16              16              0 ?   | 
                     Ar |            6               7              1 ?   | 
              Ar#c.HUtb |            7               7              0 ?   | 
              FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.HUtb |           42              42              0 ?   | 
--------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   349383
                                                      F(  8,   285) =     3.23
                                                      Prob > F      =   0.0015
Total (centered) SS     =  20027767.85                Centered R2   =   0.0127
Total (uncentered) SS   =  20027767.85                Uncentered R2 =        .
Residual SS             =  19994602.26                Root MSE      =    7.573

-----------------------------------------------------------------------------------------
                        |               Robust
              Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------------+----------------------------------------------------------------
        logpEnklavVuxen |  -.3457359   .1871668    -1.85   0.066    -.7141406    .0226689
                   intT |  -.1118093    .279168    -0.40   0.689     -.661302    .4376834
                 Kvinna |  -.1741916   .0534447    -3.26   0.001    -.2793881   -.0689952
             Gift_InvAr |   .1030048   .0657005     1.57   0.118    -.0263151    .2323246
         Barn0_15_InvAr |  -.0161521   .0201683    -0.80   0.424    -.0558498    .0235457
                        |
    c.HUtb#c.Gift_InvAr |   .0820937   .1117294     0.73   0.463    -.1378258    .3020131
                        |
        c.Kvinna#c.HUtb |   .0070085   .0821428     0.09   0.932    -.1546751     .168692
                        |
c.HUtb#c.Barn0_15_InvAr |   .0392066   .0393778     1.00   0.320    -.0383015    .1167148
-----------------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             42.900
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.1e+04
                         (Kleibergen-Paap rk Wald F statistic):         72.239
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen intT
Included instruments: Kvinna Gift_InvAr Barn0_15_InvAr c.HUtb#c.Gift_InvAr
                      c.Kvinna#c.HUtb c.HUtb#c.Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr intI
------------------------------------------------------------------------------

Absorbed degrees of freedom:
--------------------------------------------------------------------------+
            Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
------------------------+-------------------------------------------------|
               InKommun |            0             286            286 *   | 
        InKommun#c.HUtb |          285             286              1 ?   | 
                  InvAr |            4               5              1     | 
           InvAr#c.HUtb |            5               5              0 ?   | 
                IFAUkod |           15              16              1     | 
         IFAUkod#c.HUtb |           16              16              0 ?   | 
                     Ar |            6               7              1 ?   | 
              Ar#c.HUtb |            7               7              0 ?   | 
              FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.HUtb |           42              42              0 ?   | 
--------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo ca1

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    349,383    .5830278    7.613345          0        100

.   
.   lincom _b[intT]

 ( 1)  intT = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |  -.1118093    .279168    -0.40   0.689     -.661302    .4376834
------------------------------------------------------------------------------

.   lincom _b[logpEnklavVuxen]

 ( 1)  logpEnklavVuxen = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |  -.3457359   .1871668    -1.85   0.066    -.7141406    .0226689
------------------------------------------------------------------------------

.   lincom _b[logpEnklavVuxen]+_b[intT]

 ( 1)  logpEnklavVuxen + intT = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |  -.4575452   .2365742    -1.93   0.054    -.9231996    .0081093
------------------------------------------------------------------------------

.   
.   esttab hu1 hu2 hu3 hu4 using "`output'\Output\TableA8_B.rtf", keep(`controls2' logpEnklavVuxen) replace star(* .1 ** .05 *** .01) b(3) se
> (3)
(note: file C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\TableA8_B.rtf not found)
(output written to C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\TableA8_B.rtf)

.  
.   //################# Results by citizenship status  ###############################/
.    local sample "inrange(InvAr, 1987, 1991) & inrange(InvAlder, 18, 55) & inrange(Ar, 1991, 2014) & FlyktInv == 1 & Rostratt == 1 "

.   local controls1 "InvAr IFAUkod Ar FodelseAr"

.   local controls2 "Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr"

.   local controls3 "LoneInk UtbAr Sysselsatt" 

.   
.   reghdfe Nominerad (logpEnklavVuxen = logpEnklavVuxenInvAr) if SvMed == 0 & `sample', absorb(InKommun `controls1') cluster(InKommun) stage
> s(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    119,509
Absorbing 5 HDFE groups                           F(   1,    285) =     216.84
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4263
                                                  Adj R-squared   =     0.4246
                                                  Within R-sq.    =     0.1128
Number of clusters (InKommun) =        286        Root MSE        =     0.6583

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .3572362   .0242598    14.73   0.000     .3094851    .4049872
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   119509
                                                      F(  1,   285) =     3.93
                                                      Prob > F      =   0.0485
Total (centered) SS     =  3446543.832                Centered R2   =   0.0130
Total (uncentered) SS   =  3446543.832                Uncentered R2 =        .
Residual SS             =   3444317.51                Root MSE      =    5.376

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.2458778   .1240923    -1.98   0.049    -.4901314   -.0016242
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             46.927
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.5e+04
                         (Kleibergen-Paap rk Wald F statistic):        216.839
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             286            286 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo hc1

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    119,509     .292865    5.403792          0        100

.   
.   reghdfe Nominerad (logpEnklavVuxen = logpEnklavVuxenInvAr) if SvMed == 1 & `sample', absorb(InKommun `controls1') cluster(InKommun) stage
> s(first) old
(running historical version of reghdfe)
(converged in 9 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    229,874
Absorbing 5 HDFE groups                           F(   1,    285) =     112.61
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4000
                                                  Adj R-squared   =     0.3991
                                                  Within R-sq.    =     0.0442
Number of clusters (InKommun) =        286        Root MSE        =     0.6883

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .2259208   .0212895    10.61   0.000     .1840161    .2678254
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            4               5              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   229874
                                                      F(  1,   285) =     5.65
                                                      Prob > F      =   0.0181
Total (centered) SS     =  16625189.18                Centered R2   =   0.0092
Total (uncentered) SS   =  16625189.18                Uncentered R2 =        .
Residual SS             =   16591578.9                Root MSE      =    8.502

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.5913695   .2486967    -2.38   0.018    -1.080885   -.1018542
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             38.321
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.1e+04
                         (Kleibergen-Paap rk Wald F statistic):        112.611
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             286            286 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            4               5              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo hc2

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    229,874    .7338803    8.535207          0        100

.   
.   reghdfe Nominerad `controls2' (logpEnklavVuxen = logpEnklavVuxenInvAr) if SvMed == 0 & `sample', absorb(InKommun `controls1') cluster(InK
> ommun) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    119,509
Absorbing 5 HDFE groups                           F(   5,    285) =      61.07
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4273
                                                  Adj R-squared   =     0.4256
                                                  Within R-sq.    =     0.1145
Number of clusters (InKommun) =        286        Root MSE        =     0.6577

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |   .3539029   .0240358    14.72   0.000     .3065927     .401213
              Kvinna |   -.018223   .0062916    -2.90   0.004     -.030607   -.0058391
          Gift_InvAr |   .0124721   .0078097     1.60   0.111    -.0028998    .0278441
         UtbAr_InvAr |  -.0104203   .0015885    -6.56   0.000    -.0135471   -.0072936
      Barn0_15_InvAr |   .0048657   .0031266     1.56   0.121    -.0012884    .0110198
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            6               7              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   119509
                                                      F(  5,   285) =     4.31
                                                      Prob > F      =   0.0009
Total (centered) SS     =  3446543.832                Centered R2   =   0.0135
Total (uncentered) SS   =  3446543.832                Uncentered R2 =        .
Residual SS             =  3442613.922                Root MSE      =    5.375

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.2244148   .1265574    -1.77   0.077    -.4735206     .024691
         Kvinna |   -.096771   .0444852    -2.18   0.030    -.1843322   -.0092098
     Gift_InvAr |  -.0073943   .0451763    -0.16   0.870    -.0963159    .0815273
    UtbAr_InvAr |   .0398105   .0124678     3.19   0.002     .0152699    .0643511
 Barn0_15_InvAr |   .0536245   .0240741     2.23   0.027     .0062388    .1010101
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             47.279
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.5e+04
                         (Kleibergen-Paap rk Wald F statistic):        216.796
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             286            286 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            6               7              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo hc3

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    119,509     .292865    5.403792          0        100

.   
.   reghdfe Nominerad `controls2' (logpEnklavVuxen = logpEnklavVuxenInvAr) if SvMed == 1 & `sample', absorb(InKommun `controls1') cluster(InK
> ommun) stages(first) old
(running historical version of reghdfe)
(converged in 9 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    229,874
Absorbing 5 HDFE groups                           F(   5,    285) =      30.21
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4008
                                                  Adj R-squared   =     0.3999
                                                  Within R-sq.    =     0.0455
Number of clusters (InKommun) =        286        Root MSE        =     0.6878

                                     (Std. Err. adjusted for 286 clusters in InKommun)
--------------------------------------------------------------------------------------
                     |               Robust
     logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
logpEnklavVuxenInvAr |    .223745   .0211163    10.60   0.000     .1821813    .2653088
              Kvinna |  -.0070761   .0060563    -1.17   0.244    -.0189969    .0048447
          Gift_InvAr |   .0137924   .0093113     1.48   0.140    -.0045352      .03212
         UtbAr_InvAr |  -.0093907   .0012492    -7.52   0.000    -.0118496   -.0069319
      Barn0_15_InvAr |   .0040576   .0032708     1.24   0.216    -.0023804    .0104957
--------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------------------------+
         Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
---------------------+-------------------------------------------------|
            InKommun |            0             286            286 *   | 
               InvAr |            4               5              1     | 
             IFAUkod |           15              16              1     | 
                  Ar |            4               5              1 ?   | 
           FodelseAr |           41              42              1 ?   | 
-----------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   229874
                                                      F(  5,   285) =    20.48
                                                      Prob > F      =   0.0000
Total (centered) SS     =  16625189.18                Centered R2   =   0.0108
Total (uncentered) SS   =  16625189.18                Uncentered R2 =        .
Residual SS             =  16565339.35                Root MSE      =    8.496

---------------------------------------------------------------------------------
                |               Robust
      Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
logpEnklavVuxen |  -.4946825   .2484778    -1.99   0.047     -.983767    -.005598
         Kvinna |  -.2344789   .0516497    -4.54   0.000    -.3361421   -.1328157
     Gift_InvAr |   .2674892   .0814836     3.28   0.001     .1071033    .4278751
    UtbAr_InvAr |   .1197318     .01445     8.29   0.000     .0912895     .148174
 Barn0_15_InvAr |  -.0422028   .0254836    -1.66   0.099    -.0923627    .0079572
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             38.358
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):              1.0e+04
                         (Kleibergen-Paap rk Wald F statistic):        112.272
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen
Included instruments: Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
Excluded instruments: logpEnklavVuxenInvAr
------------------------------------------------------------------------------

Absorbed degrees of freedom:
------------------------------------------------------------------+
    Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
----------------+-------------------------------------------------|
       InKommun |            0             286            286 *   | 
          InvAr |            4               5              1     | 
        IFAUkod |           15              16              1     | 
             Ar |            4               5              1 ?   | 
      FodelseAr |           41              42              1 ?   | 
------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo hc4

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    229,874    .7338803    8.535207          0        100

.   
.   // Code to significance the difference, use a fully interacted model to replicate the split sample results 
.   capture drop intT intI

.   gen intT = logpEnklavVuxen*SvMed

.   gen intI = logpEnklavVuxenInvAr*SvMed

.   
.   reghdfe Nominerad  `controls2' c.SvMed#c.Gift_InvAr c.Kvinna#c.SvMed c.SvMed#c.Barn0_15_InvAr c.SvMed#c.UtbAr_InvAr (logpEnklavVuxen intT
>  = logpEnklavVuxenInvAr intI) if `sample', absorb(InKommun##c.SvMed InvAr##c.SvMed IFAUkod##c.SvMed Ar##c.SvMed FodelseAr##c.SvMed) cluster
> (InKommun) stages(first) old
(running historical version of reghdfe)
(converged in 10 iterations)

Stage: first - logpEnklavVuxen

HDFE Linear regression                            Number of obs   =    349,383
Absorbing 5 HDFE groups                           F(  10,    285) =      41.45
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.4336
                                                  Adj R-squared   =     0.4324
                                                  Within R-sq.    =     0.0688
Number of clusters (InKommun) =        286        Root MSE        =     0.6777

                                             (Std. Err. adjusted for 286 clusters in InKommun)
----------------------------------------------------------------------------------------------
                             |               Robust
             logpEnklavVuxen |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-----------------------------+----------------------------------------------------------------
        logpEnklavVuxenInvAr |   .3539029   .0240435    14.72   0.000     .3065775    .4012282
                        intI |  -.1301578    .008772   -14.84   0.000    -.1474239   -.1128918
                      Kvinna |   -.018223   .0062936    -2.90   0.004    -.0306109   -.0058352
                  Gift_InvAr |   .0124721   .0078122     1.60   0.111    -.0029048     .027849
                 UtbAr_InvAr |  -.0104203    .001589    -6.56   0.000    -.0135481   -.0072926
              Barn0_15_InvAr |   .0048657   .0031276     1.56   0.121    -.0012904    .0110218
                             |
    c.SvMedborg#c.Gift_InvAr |   .0013203   .0095948     0.14   0.891    -.0175653    .0202058
                             |
        c.Kvinna#c.SvMedborg |   .0111469    .006215     1.79   0.074    -.0010863    .0233801
                             |
c.SvMedborg#c.Barn0_15_InvAr |  -.0008081   .0031488    -0.26   0.798     -.007006    .0053898
                             |
   c.SvMedborg#c.UtbAr_InvAr |   .0010296   .0014963     0.69   0.492    -.0019155    .0039747
----------------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------------------+
                 Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
-----------------------------+-------------------------------------------------|
                    InKommun |            0             286            286 *   | 
        InKommun#c.SvMedborg |          286             286              0 ?   | 
                       InvAr |            4               5              1     | 
           InvAr#c.SvMedborg |            5               5              0 ?   | 
                     IFAUkod |           15              16              1     | 
         IFAUkod#c.SvMedborg |           16              16              0 ?   | 
                          Ar |            6               7              1 ?   | 
              Ar#c.SvMedborg |            5               7              2 ?   | 
                   FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.SvMedborg |           42              42              0 ?   | 
-------------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation
WARNING: Missing F statistic (dropped variables due to collinearity or too few clusters).

Stage: first - intT

HDFE Linear regression                            Number of obs   =    349,383
Absorbing 5 HDFE groups                           F(   6,    285) =          .
Statistics robust to heteroskedasticity           Prob > F        =          .
                                                  R-squared       =     0.9386
                                                  Adj R-squared   =     0.9385
                                                  Within R-sq.    =     0.0455
Number of clusters (InKommun) =        286        Root MSE        =     0.5581

                                             (Std. Err. adjusted for 286 clusters in InKommun)
----------------------------------------------------------------------------------------------
                             |               Robust
                        intT |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-----------------------------+----------------------------------------------------------------
        logpEnklavVuxenInvAr |   6.82e-16          .        .       .            .           .
                        intI |    .223745   .0211262    10.59   0.000     .1821619    .2653282
                      Kvinna |  -4.12e-17   1.73e-10    -0.00   1.000    -3.40e-10    3.40e-10
                  Gift_InvAr |  -2.84e-16   9.19e-10    -0.00   1.000    -1.81e-09    1.81e-09
                 UtbAr_InvAr |   7.53e-17          .        .       .            .           .
              Barn0_15_InvAr |  -5.63e-18   2.12e-10    -0.00   1.000    -4.18e-10    4.18e-10
                             |
    c.SvMedborg#c.Gift_InvAr |   .0137924   .0093156     1.48   0.140    -.0045438    .0321285
                             |
        c.Kvinna#c.SvMedborg |  -.0070761   .0060591    -1.17   0.244    -.0190025    .0048502
                             |
c.SvMedborg#c.Barn0_15_InvAr |   .0040576   .0032724     1.24   0.216    -.0023834    .0104987
                             |
   c.SvMedborg#c.UtbAr_InvAr |  -.0093907   .0012498    -7.51   0.000    -.0118508   -.0069307
----------------------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------------------+
                 Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
-----------------------------+-------------------------------------------------|
                    InKommun |            0             286            286 *   | 
        InKommun#c.SvMedborg |          286             286              0 ?   | 
                       InvAr |            4               5              1     | 
           InvAr#c.SvMedborg |            5               5              0 ?   | 
                     IFAUkod |           15              16              1     | 
         IFAUkod#c.SvMedborg |           16              16              0 ?   | 
                          Ar |            6               7              1 ?   | 
              Ar#c.SvMedborg |            5               7              2 ?   | 
                   FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.SvMedborg |           42              42              0 ?   | 
-------------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

Stage: iv

HDFE IV (2SLS) estimation
-------------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on InKommun

Number of clusters (InKommun) =    286                Number of obs =   349383
                                                      F( 10,   285) =    10.66
                                                      Prob > F      =   0.0000
Total (centered) SS     =  20071733.01                Centered R2   =   0.0120
Total (uncentered) SS   =  20071733.01                Uncentered R2 =        .
Residual SS             =  20007953.27                Root MSE      =    7.575

----------------------------------------------------------------------------------------------
                             |               Robust
                   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-----------------------------+----------------------------------------------------------------
             logpEnklavVuxen |  -.2244148   .1265981    -1.77   0.077    -.4736007     .024771
                        intT |  -.2702677   .2375158    -1.14   0.256    -.7377754      .19724
                      Kvinna |   -.096771   .0444995    -2.17   0.030    -.1843604   -.0091817
                  Gift_InvAr |  -.0073943   .0451909    -0.16   0.870    -.0963445    .0815559
                 UtbAr_InvAr |   .0398105   .0124718     3.19   0.002      .015262     .064359
              Barn0_15_InvAr |   .0536245   .0240819     2.23   0.027     .0062236    .1010253
                             |
    c.SvMedborg#c.Gift_InvAr |   .2748835   .0906187     3.03   0.003     .0965168    .4532503
                             |
        c.Kvinna#c.SvMedborg |  -.1377079   .0599301    -2.30   0.022    -.2556697   -.0197462
                             |
c.SvMedborg#c.Barn0_15_InvAr |  -.0958272   .0325626    -2.94   0.004    -.1599209   -.0317336
                             |
   c.SvMedborg#c.UtbAr_InvAr |   .0799212   .0154571     5.17   0.000     .0494966    .1103458
----------------------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             38.358
                                                   Chi-sq(1) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):             7887.315
                         (Kleibergen-Paap rk Wald F statistic):         56.083
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         logpEnklavVuxen intT
Included instruments: Kvinna Gift_InvAr UtbAr_InvAr Barn0_15_InvAr
                      c.SvMedborg#c.Gift_InvAr c.Kvinna#c.SvMedborg
                      c.SvMedborg#c.Barn0_15_InvAr c.SvMedborg#c.UtbAr_InvAr
Excluded instruments: logpEnklavVuxenInvAr intI
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-------------------------------------------------------------------------------+
                 Absorbed FE |  Num. Coefs.  =   Categories  -   Redundant     | 
-----------------------------+-------------------------------------------------|
                    InKommun |            0             286            286 *   | 
        InKommun#c.SvMedborg |          286             286              0 ?   | 
                       InvAr |            4               5              1     | 
           InvAr#c.SvMedborg |            5               5              0 ?   | 
                     IFAUkod |           15              16              1     | 
         IFAUkod#c.SvMedborg |           16              16              0 ?   | 
                          Ar |            6               7              1 ?   | 
              Ar#c.SvMedborg |            5               7              2 ?   | 
                   FodelseAr |           41              42              1 ?   | 
       FodelseAr#c.SvMedborg |           42              42              0 ?   | 
-------------------------------------------------------------------------------+
? = number of redundant parameters may be higher
* = fixed effect nested within cluster; treated as redundant for DoF computation

.   eststo ca1

.   sum Nominerad if e(sample)==1

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
   Nominerad |    349,383    .5830278    7.613345          0        100

.   
.   lincom _b[intT]

 ( 1)  intT = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |  -.2702677   .2375158    -1.14   0.256    -.7377754      .19724
------------------------------------------------------------------------------

.   lincom _b[logpEnklavVuxen]

 ( 1)  logpEnklavVuxen = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |  -.2244148   .1265981    -1.77   0.077    -.4736007     .024771
------------------------------------------------------------------------------

.   lincom _b[logpEnklavVuxen]+_b[intT]

 ( 1)  logpEnklavVuxen + intT = 0

------------------------------------------------------------------------------
   Nominerad |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         (1) |  -.4946825   .2485937    -1.99   0.048    -.9839952   -.0053698
------------------------------------------------------------------------------

.   
.   esttab hc1 hc2 hc3 hc4 using "`output'\Output\TableA8_C.rtf", keep(`controls2' logpEnklavVuxen) replace star(* .1 ** .05 *** .01) b(3) se
> (3)
(note: file C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\TableA8_C.rtf not found)
(output written to C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\TableA8_C.rtf)

.  
.   log close 
      name:  <unnamed>
       log:  C:\Userdata\Shared\Dofiles\DoAnalysis\PlacementCandidacy\Output\Fig3_TabA6_A7.log
  log type:  text
 closed on:   2 Oct 2020, 10:42:27
---------------------------------------------------------------------------------------------------------------------------------------------
